An International Comparison of Capital Structure and Debt Maturity Choices* Joseph P.H. Fan† Faculty of Business Administration The Chinese University of Hong Kong Shatin, N.T. Hong Kong Email: [email protected]

Sheridan Titman McCombs School of Business University of Texas at Austin Austin, TX 78712 USA Email: [email protected] and

Garry Twite School of Finance and Applied Statistics Australian National University Canberra, ACT 0200 Australia Email: [email protected]

*

This paper has benefited from the useful comments and suggestions provided by Andres Almazan, Heitor

Almeida, Lawrence Booth, Stijn Claessens, Joshua Coval, Sudipto Dasgupta, Jay Hartzell, Jiang Luo, Vojislav Maksimovic, Enrico Perotti, Tom Smith, participants at the 2003 European Finance Association Conference, the 2003 Financial Management Association Conference and the 2005 American Finance Association Meeting, seminar participants at the Australian National University, Australian Graduate School of Management, Chinese University of Hong Kong, Hong Kong University of Science and Technology, Shanghai University of Finance & Economics, University of Melbourne, University of Queensland, University of Sydney, and University of Texas at Austin, and anonymous referees. Joseph Fan thanks the financial support from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project No. CUHK6230/03H) and the University of Queensland for research support during his visit when part of the research was carried out. Garry Twite thanks the financial support from the Australian Research Council Discovery Project (Project ID. DP0664505). †

Corresponding author. Faculty of Business Administration, Chinese University of Hong Kong, Shatin,

N.T., Hong Kong; email: [email protected]; phone:+ 852 2609-7839; fax: +852 2603-5114

Abstract This study examines the influence of institutional environment on capital structure and debt maturity choices by examining a cross-section of firms in 39 developed and developing countries. We find that a country’s legal and tax system, the level of corruption and the preferences of capital suppliers explain a significant portion of the variation in leverage and debt maturity ratios. Our evidence indicate that firms in countries that are viewed as more corrupt tend to use less equity and more debt, especially short-term debt, while firms operating within legal systems that provide better protection for financial claimants tend to have capital structures with more equity, and relatively more long-term debt. In addition, the existence of an explicit bankruptcy code and/or deposit insurance is associated with higher leverage and more long-term debt. We also find that firms tend to use more debt in countries where there is a greater tax gain from leverage, while firms in countries with larger government bond markets have lower leverage, suggesting that government bonds tend to crowd out corporate debt. Countries with more extensive defined benefit pension funds have higher debt ratios and longer debt maturities, whereas those with more extensive defined contribution fund activities have lower debt ratios. In addition, debt ratios are lower in countries that limit the bond holdings of pension funds. Finally, we do not find a significant association between financing choices and the size of the insurance industry.

I. Introduction Corporate financing choices are determined by a combination of factors that are related to the characteristics of the firm as well as to their institutional environment. Although most studies focus on the importance of firm characteristics by examining corporate financing choices within individual countries, 1 there is a growing literature that considers how institutional differences affect these choices. To explore the cross-sectional variation in the institutional environment, these papers examine capital structure choices across countries (Booth, Aivazian, Demirguc-Kunt, and Maksimovic (2001), Claessens, Djankov, and Nenova (2001), Demirguc-Kunt and Maksimovic (1996), Demirguc-Kunt and Maksimovic (1998), Demirguc-Kunt and Maksimovic (1999), Giannetti (2003), De Jong, Kabir, and Nguyen (2008)). This study builds on this recent literature in two important ways. First, because we consider these issues within a panel that includes industry fixed effects, together with firm-level variables, we identify the variation in capital structure across countries that cannot be explained by cross-country differences in the industrial mix and firm-level characteristics. Second, we consider a larger number of countries and a number of important institutional characteristics not previously explored in this literature. To understand our motivation, it is useful to illustrate the importance of countrylevel factors relative to industry factors in determining capital structure. A regression of 1

Examples of empirical studies examining the association between firm characteristics and capital

structure within specific countries include Titman and Wessels (1988) – U.S., Campbell and Hamao (1995) – Japan and Gatward and Sharpe (1996) – Australia. Barclay and Smith (1995), Stohs and Mauer (1996) and Guedes and Opler (1996) examine the association between firm characteristics and debt maturity in the U.S. Gatward and Sharpe (1996) undertake a similar study of debt maturity in Australia.

1

firm leverage, measured as the book value of debt over the market value of the firm, on firm-specific variables, industry fixed effects and country fixed effects, has an adjusted R-square of 0.19. When the regression is estimated with all variables except for country fixed effects, the adjusted R-square is reduced to 0.15.2 However, in a regression that includes all variables except for industry dummies the adjusted R-square is reduced by only half as much, to 0.17. When the full regression is estimated with debt maturity, measured as the book value of long-term debt to the book value of total debt, as the dependent variable, the R-square is 0.25. When the regression is estimated with all variables except for country fixed effects, the R-square is substantially reduced to 0.09. However, in the regression that includes all variables except for industry fixed effects, the R-square is only slightly reduced to 0.23. These experiments indicate that the country in which the firm resides is a more important determinant of how it is financed than is its industry affiliation, which in turn suggests that differences in country level institutional factors are likely to have a first order effect on capital structure choices. To examine this possibility in more detail, we estimate a panel regression on a large sample of firms from 39 countries that examines the extent to which cross-country differences in capital structures can be explained by differences in tax policies; legal environment; and the importance and regulation of financial institutions. Our evidence suggests that the explanatory power of a model that includes between ten and twelve institutional variables explains the cross-section of debt ratios

2

This result is similar in character to a regression reported by Booth, Aivazian, Demirguc-Kunt, and

Maksimovic (2001).

2

and maturity structures about as well as a model with 39 country dummy variables. These regressions indicate that firms tend to use more debt in countries with a greater tax gain from leverage, which contrasts with Booth, Aivazian, Demirguc-Kunt, and Maksimovic (2001) who, in an earlier study of mostly developing economies, do not find a significant relation between debt ratios and tax policy. In addition, we find that the strength of a country’s legal system and public governance importantly affect firm capital structure. Weaker laws and more government corruption are associated with higher corporate debt ratios and shorter debt maturity.3 We also find that countries with deposit insurance or explicit bankruptcy codes, like the Chapter 11 and Chapter 7 rules in the U.S., have higher debt ratios and longer debt maturities. These findings reinforce the prior literature on the importance of the legal system, the enforcement of investor rights and financial distress resolution (Claessens, Djankov, and Mody (2001), Djankov, Hart, McLiesh, and Shleifer (2008), La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1997), La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1998)). We also find that the preferences of the suppliers of capital influence capital structure choices. 4 In particular, firms in countries with larger banking sectors have shorter maturity debt, but the association between financing choices and the size of the insurance industry is weak. In addition, firms in countries with higher levels of defined contribution pension fund assets use relatively more equity, while firms in countries with 3

This result is consistent with Demirguc-Kunt and Maksimovic (1999).

4

One should interpret these results with some caution, because an analysis of capital suppliers does raise

endogeneity concerns. In particular, we expect financial institutions to develop in ways that satisfy the financing needs of firms. However, as discussed in Section 2.3, we have selected variables that are less likely to be influenced by the capital structure preferences of corporations.

3

higher levels of defined benefit pension fund assets use relatively more long-term debt, which could reflect differences in how these pension assets are invested. Finally, we find that firms in countries with larger government bond markets have lower debt ratios and shorter maturity debt, indicating that government bonds tend to crowd out long term corporate debt. The paper is organized as follows: Section 2 discusses the association between country level institutional factors and financial choices. Section 3 introduces the set of firm level variables that influence capital structure choice.

Section 4 describes the

sample. Section 5 presents our results and Section 6 draws some conclusions.

II. Institutional Factors and Cross-Country Determinants of Capital Structure This section discusses how institutional differences between countries can potentially affect how firms within these countries are financed. Specifically, we consider institutional variables that reflect (1) the ability of creditors to enforce legal contracts (2) the tax treatment of debt and equity, and (3) the importance and regulation of financial institutions that represent major suppliers of capital. We expect that weaker legal systems and weaker public enforcement of laws should be associated with less external equity and shorter maturity debt contracts. We also expect that firms in countries with lower tax preferences for debt will be less levered. Finally, we examine whether the suppliers of capital matter. Although most of the capital structure literature focuses on the financing preferences of firms, at the aggregate level, firm capital structures are determined by the preferences of the suppliers of capital (i.e. investors) as well as the preferences of firms. In particular, exogenous factors that lead

4

suppliers of capital to prefer to hold more or less equity relative to debt will also influence the capital structures of firms. The following sub-sections introduce the variables that we consider, and discuss how these variables are likely to influence typical debt ratios within our sample of countries.

A. Legal System Incentive problems - conflicts of interest between corporate insiders (managers, employees and/or majority shareholders) and external investors - are important factors that shape corporate policy and productivity. As extensively explored by La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1998), the extent to which contracts can be used to mitigate these problems depends on the legal system, which consists of both the content of the laws and the quality of their enforcement. In the following discussion we will review how these legal factors influence financing choices. In countries with weak laws and enforcement, financial instruments (e.g. short term debt) that allow insiders less discretion, and are contractually easier to interpret, are likely to dominate.

La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1998) find

significant variation in the extent of legal protection of external investors across both developed and developing countries, and argue that legal systems based on common law offer outside investors (debt and equity) better protection than those based on civil law, resulting in higher security values (La Porta, Lopez-de-Silanes, Shleifer, and Vishny (2002)). All else equal, this suggests that common law countries will use more outside equity and longer-term debt. To test whether this is the case, we define an indicator

5

variable that takes a value of one if the country’s legal system is based on common law and zero otherwise. In addition to the content of the law, the integrity and enforceability of the law is also important, which we measure by the perceived corruption level in a country. Corruption has been identified as a key factor shaping a country’s legal system (Djankov, La Porta, Lopez-de-Silanes, and Shleifer (2003)), resource allocation and firm behavior (La Porta, Lopez-De-Silanes, Shleifer, and Vishny (1999), Fisman (2001), Johnson and Mitton (2003)). We are not the first to examine the roles of legal factors in corporate financing choices. Demirguc-Kunt and Maksimovic (1999) find that firms have longer duration debt in countries where the legal system has more “integrity”. Integrity, measured by a law and order index prepared by the International Country Risk Guide, reflects the extent to which individuals are willing to rely on the legal system to make and implement laws, mediate disputes and enforce contracts. In contrast, we focus on corruption, defined as the abuse of public office for private gain, measured as the Corruption Perception Index (Transparency International), which reflects the extent to which corruption is perceived to exist among public officials and politicians. An advantage of this index is that it provides both time-series and cross-sectional variation; most other measures of integrity, such as the law and order index, do not have comparable historical data. We reverse the index, so that in our study, it ranges from 0 to 10, with larger values indicating more severe corruption. In the context of the firm’s capital structure choices, the index proxies for the threat of all or part of investor rights being expropriated by managers or public officials. Debt is expected to be used relatively more than equity

6

when the public sector is more corrupt, since it is easier to expropriate outside equity holders than debt holders.

Similarly, one can argue that since short-term debt is more

difficult to expropriate, it will be used relatively more frequently than long-term debt in more corrupt countries. Finally, we investigate the enforcement of debt contracts. As identified by Djankov, Hart, McLiesh, and Shleifer (2008), the legal structure that specifies the resolution of default differs widely across countries. Indeed, in some countries, like the U.S., there is an explicit bankruptcy code that specifies and limits the rights and claims of creditors that facilitates the reorganization of the ongoing business. In contrast, in other countries with no bankruptcy codes or only weakly enforced codes, creditors often have difficulty accessing collateral by liquidating distressed firms or seizing distressed firm assets (Claessens, Djankov, and Mody (2001), Claessens, Djankov, and Klapper (2003), Claessens and Klapper (2005), Davydenko and Franks (2008), Dinc (2005)). With poorly defined bankruptcy procedures, senior lenders typically possess inferior bargaining power relative to the borrower in out-of-court renegotiations due to the weak laws and lenders’ inherent information disadvantage about the collateral relative to borrowers (Degryse and Ongena (2005), Petersen and Rajan (1994)) lowering demand for long-term debt. On the borrower side, the existence of defined bankruptcy procedures for corporate reorganization and the deferral of debt payments increase the incentive for firms to issue long term debt since a default can be less onerous.5 We conjecture that the lack of explicit bankruptcy codes and enforcement discourage the use of debt, in particular long-term

5

The influence of the existence of a bankruptcy code on both investor demand for, and corporate supply of

long-term debt was pointed out to the authors by an anonymous referee.

7

debt. Based on Djankov, Hart, McLiesh, and Shleifer (2008), we define an indicator variable that takes a value of one for those countries in which an insolvent firm can undergo a court supervised reorganization proceeding.

B. Tax Code The tax system in general, and specifically the tax treatment of interest and dividend payments, has been recognized as an important factor influencing capital structure choices since the seminal work of Modigliani and Miller (1963).6 We observe three main categories of tax regimes: The first is the classical tax system in which dividend payments are taxed at both the corporate and personal levels and interest payments are tax-deductible corporate expenses. The classical tax system exists in Brazil, Chile, China, Hong Kong, India, Indonesia, Israel, Japan, Korea, Malaysia, Netherlands, Pakistan, Peru, Philippines, Singapore, South Africa, Switzerland, United Kingdom (post 2000) 7 and the United States. The second is the dividend relief tax system, where dividend payments are taxed at a reduced rate at the personal level. A dividend relief tax system exists in Austria, Belgium, Denmark, Greece, Portugal, Sweden, Thailand and Turkey.8 In Greece and Turkey dividend payments are not taxed at the personal level, that is, a full dividend relief system. 6

See Graham (2003) for a review of the literature on the influence of taxes on capital structure choice.

7

The United Kingdom reverted to a classical tax system in 2001.

8

The United States currently provides preferential tax treatment for dividend over interest payments, but

not in our sample period.

8

Third is the dividend imputation tax system, where corporations can deduct interest payments, but where the domestic shareholders of a corporation receive a tax credit for the taxes paid by the corporation. The goal of the system is to tax corporate profits only once. Dividend imputation systems are in place in Australia, Canada, France, Germany, Ireland, Italy, Mexico, New Zealand, Norway, Spain, Taiwan and United Kingdom (pre 2001). The proportion of corporate tax available as a tax credit under these imputation systems varies from country to country. In Australia, Finland, Germany, Italy, New Zealand and Norway the full amount of the corporate tax paid is distributed as a tax credit. In other countries only part of the corporate tax credits are distributed. All else equal, we expect that debt will be used less in countries with dividend imputation or tax relief systems than in countries with classical tax systems that double tax corporate profits. To test for this relationship for each country in our sample, we estimate the tax shield, using the tax gain from leverage variable introduced in Miller (1977):

1−

(1 − τ c )(1 − τ e ) (1 − τ i )

where τ c is the statutory corporate tax rate, τ i is the highest statutory personal tax rate on interest income and τ e is the highest effective personal tax rate on equity income coming from dividends.9

9

We also consider a dividend tax indicator variable that takes a value of one for countries with either a full

dividend relief tax system or a full dividend imputation tax system and zero otherwise.

9

The tax gain from leverage can take both positive and negative values. Negative values arise under a dividend relief tax system, when the personal tax rate on interest income is greater than the corporate tax rate and the personal tax rate on dividend income is less than the corporate tax rate. This is the case under a full dividend relief system as exists in Greece and Turkey, as well as under some partial relief countries like Belgian and Thailand The tax gain from leverage is zero under a full dividend imputation tax system, which is the case in Australia, Germany, Italy, New Zealand and Norway. For all other countries the value of the tax gain from leverage is positive.

C. Suppliers of Capital Financial economists have typically viewed the capital structure problem from the perspective of firms that face competitive and complete financial markets, where debt and equity capital are offered at equivalent risk-adjusted rates. However, when this is not the case, the preferences of investors to hold debt versus equity instruments will have an influence on how firms are financed. For example, in the Miller (1977) model, the aggregate debt ratio in the economy is determined by aggregate investor preferences for holding debt versus equity securities. While these preferences are determined by taxes in Miller’s model, one can more generally consider how investor preferences for holding various debt and equity instruments affect the capital structure choice of firms.10 We will specifically consider the preferences of banks, pension funds and insurance companies. Banks tend to have short-term liabilities and thus may have a comparative advantage holding short-term debt. In contrast, pension funds have long-

10

See Titman (2002) for a discussion of the effect of investor preferences on capital structure choices.

10

term liabilities, and thus have a preference for holding long term assets. Likewise, insurance companies tend to hold longer term assets. Hence, we might expect firms in counties with a larger banking sector to use more short-term financing and firms in countries with larger pension funds and insurance sectors to use more long-term financing. The analysis of supply effects raise endogeneity concerns, since we expect financial intermediaries to develop in ways that satisfy the financing needs of firms as well as the preferences of investors. Existing studies (for example, Dermirguc-Kunt and Maksimovic (1999), De Jong, Kabir, and Nguyen (2008)) examine the effects of stock/bond market size, turnover and bank total assets on capital structure choices. These variables, however, are likely to be influenced by the capital structure preferences of corporations. For example, in countries with industries (like high tech) that require considerable amounts of external capital, the stock market is likely to be larger.11 With this in mind, we depart from the existing literature and select proxies that are not likely to be directly influenced by the capital structure preferences of corporations. In particular, we select measures of the supply of funds available to these financial intermediaries.

11

Demirguc-Kunt and Maksimovic (1999) recognise this endogeneity issue and address it by using a two-

stage instrumental variable regression. They chose as instruments measures of the size of the economy and the flow of funds, plus proxies for the content, strength and integrity of the legal system. However, one can argue that these variables either directly influence the capital structure choice or are potentially influenced by the types of firms in the economy, and are thus indirectly related to the capital structure choice.

11

To proxy for the supply of funds to banks, we use deposits/GDP to measure the amount of funds that are available to the banking sector.12 In addition, deposit insurance is used in many countries to protect bank depositors, in full or in part, from losses caused by a bank's inability to pay its debts when due. This lowers the risk of bank runs, reducing the banks need to hedge and seek more liquid short-term debt. We conjecture that the existence of deposit insurance will influence the lending and maturity choices of banks. In particular, banks provide more credit to firms and lend longer term debt when deposits are protected. Hence, one might expect that firms in countries with deposit insurance to have higher leverage and more long-term debt. To test this relationship, we utilize a deposit insurance indicator variable that takes a value of one if bank deposits are at least partially explicitly insured by government and zero otherwise (Demirguc-Kunt, Karacaovali, and Laeven (2005)). We use insurance penetration (value of total insurance premiums (life and nonlife)/GDP) to measure the amount of funds that are available to insurance companies. Different insurance companies, however, may have very different liability structures and

12

It is possible that there are unobserved factors that affect both the willingness of investors to deposit

funds with banks and the willingness of banks to provide long-term funding to firms, creating a spurious relation between deposits and capital structure. For example, one can argue that the financing needs of corporations affect the funds that are available to the different investor sectors. Suppose, for example, that the need for monitoring declines, making bank loans somewhat less attractive to long-term bonds. On the margin, this would increase the interest rate on long-term bonds, making it more attractive for households to invest in fixed income mutual funds rather than bank deposits.

While this creates a potential

endogeniety problem, it is mitigated by the inclusion of our institutional variables and probably has a minor influence on our estimates.

12

may thus have different preferences for the assets that they hold. For example, life insurance companies that offer contracts with a substantial savings component, such as whole life contracts, might have a preference for long term debt. In contrast, insurance companies that offer term life and property and casualty insurance tend to have shorterterm obligations, and thus, are expected to hold shorter-term debt. Unfortunately, we do not have data that allows us to distinguish between the different sectors of the insurance industry. We measure pension fund penetration separately as the value of defined benefit pension fund assets over GDP and the value of defined contribution pension fund assets over GDP. This distinction may be important because in firms with defined benefit plans, the asset allocation is determined by the plan sponsors, while with defined contribution plans, the asset allocation is made by individuals. It should also be noted that defined benefit pension plans are often underfunded, creating a liability that we do not observe in our data set. In addition, since it is possible that cross-country differences in pension fund regulations influence the investment choices of pension funds, we also examine restrictions on debt and equity holdings of pension funds. We conjecture that the relative restrictions on debt and equity holdings will influence the choice between debt and equity. In particular, pension funds will hold more equity when restrictions on bond holdings are tighter relative to those on equity holdings. Hence, one might expect that firms in countries with tighter restriction on bond holdings to issue more equity. To investigate this possibility we estimate an index of relative restrictions on debt and equity holdings measured as the ratio of the proportional limit on equity holdings over the proportional

13

limit on debt holdings taken from the Survey of Investment Regulation of Pension Funds, OECD. For each year we ranked countries by their pension fund regulation index, assigning countries into quartiles. We assigned a score of 1 to 4 to the quartiles, with larger values indicating tighter restriction on bond holdings. An alternative measure of the supply of funds available to financial intermediaries is the level of domestic savings, which we measure as gross domestic saving over GDP. In addition, we examine the size of the government bond market in each country by including domestically denominated government bonds/GDP as an independent variable. Government bonds can influence the supply of debt capital that is available to the corporate market for two reasons. The first is a simple crowding out argument. If there is a fixed supply of debt capital, then government debt can compete for that fixed supply and leave less available for corporate borrowers. The second possibility is that the supply is not fixed, and that the presence of government borrowers can help the debt market develop, increasing the demand for corporate debt.

III. Firm Level Characteristics and Capital Structure Choice Consistent with the existing literature (Titman and Wessels (1988), Guedes and Opler (1996), Rajan and Zingales (1995)) we include a set of firm level variables that capture factors that are known to affect leverage and maturity structure. These variables include asset tangibility (fixed assets over total assets), profitability (net income over total assets), firm size (natural logarithm of total assets) and the market-to-book ratio (market value of equity over book value of equity). Due to data limitations in some of the countries included in our study, we do not include variables that measure the effective

14

tax rate, operating risk, research and development expenditure, capital expenditure and selling expenses as per Titman and Wessels (1988). In place of these variables we include the market-to-book ratio, which can proxy for growth as well as the collateral value of assets, and industry indicator variables based on two-digit SIC codes.13

IV. Data and Sample This section describes the sample and presents the country and industry patterns of capital and debt maturity structures. It then introduces the empirical procedure employed in this study.

A. Sample Selection The primary source of our firm-level data is Worldscope, which contains financial data on companies from a wide range of industries in over 50 countries. We restrict the sample to those firms listed on the stock market of the country in which it is domiciled. Our analysis covers the period of 1991 through 2006. We exclude firm-year observations with missing financial data that is required for the firm-level analysis. The final sample consists of 36,767 firms from 39 countries, totalling 272,092 firm-years. Table 1 provides a description of the sample, which covers a broad cross section of developed and developing countries with every continent represented.

Most of the

countries have observations in each of the 16 years. As can be seen from the last two columns of Table 1, the coverage of the sample firms varies across countries in terms of number and/or market capitalization, reflecting 13

See MacKay and Phillips (2005) for evidence on the importance of industry fixed effects.

15

that Worldscope has uneven coverage of firms across the countries. For most of the economies we have more than 60 percent of sample coverage in terms of market capitalization and 50 percent in terms of number of listed firms. The economies with lower data coverage tend to be developing economies. [Table 1 about here]

B. Country Financing Patterns Our measures of capital structure are: (i)

leverage, measured as the proportion of total debt to market value of the firm (total debt/market value). Total debt is defined to be the book value of shortterm and long-term interest bearing debt. Market value of the firm is defined as the market value of common equity plus book value of preferred stock plus total debt, or

(ii)

debt maturity, measured as the proportion of the book value of long-term debt to total debt (long-term debt/total debt).14

To gain a basic idea about how capital and maturity structures differ across countries, we compute the median leverage and maturity structure by country for the period 1991 to 2006. As can be seen in Figure 1, developing economies occupy both ends of the leverage spectrum.

The highest five leverage ratios are observed in South Korea,

Indonesia, Brazil, Portugal, and Pakistan, while the lowest five are observed in Australia,

14

Trade credit is an important source of financing in economies with underdeveloped financial institutions

(Demirguc-Kunt and Maksimovic (2001), Fisman and Love (2003)). Our results are robust to including trade credit (measured as accounts payable) in our measure of short-term debt.

16

South Africa, Canada, the United States, and Turkey. Developing economies seem to dominate the higher range, while developed economies tend to be at the lower range. The median leverage ratio for the developing economies in the sample is 0.26,15 while for the developed economies the median leverage ratio is 0.20. The middle range of the leverage spectrum is mixed with both developing and developed economies. [Figure 1 about here] Figure 2 presents the median maturity structure by country. It is clear from the figure that debt obligations have longer maturities in more developed economies. The five countries with the highest long-term debt ratios are New Zealand, Norway, Sweden, USA, and Canada. The lowest five median long-term debt ratios are observed in China, Greece, Turkey, Taiwan, and Thailand. 16 The median long-term debt ratio for the developing economies in the sample is 0.36, while for the developed economies the median long-term debt ratio is 0.61. [Figure 2 about here] In addition to the set of firm and country-level variables described in Section 2, we include inflation, inflation volatility (measured as the standard deviation of inflation rates over the preceding four years) and a developed economy indicator variable that takes a value of one if the country is classified as a developed economy according to the

15

Economies within the sample classified as developing, according to the World Bank, are Brazil, Chile,

China, India, Indonesia, Malaysia, Mexico, Pakistan, Peru, Philippines, Taiwan, Thailand, Turkey and South Africa. 16

This parallels the findings of Demirguc-Kunt and Maksimovic (1999) for an early sample period, 1980-

1991.

17

World Bank classification that is based on the countries’ gross national income levels.17 Inflation is included because debt contracts are generally nominal contracts and high inflation, which is generally associated with high uncertainty about future inflation, may tilt lenders away from long-term debt. Likewise, higher inflation volatility reflects higher uncertainty about future inflation, tilting lenders away from long-term debt. A developed economy indicator variable is included because it may pick up an element of economic development that is not already captured by our other variables. Both firm and country level variables are lagged one period to allow for the non-contemporaneous nature of the interaction between firm/country level characteristics and financing choices. Table 2, which presents the summary statistics, shows cross-sectional variation in the country-level variables. The country-level variables are defined in Table A1, along with their data sources. Except for the common law, developed economy, bankruptcy code and deposit insurance variables that are constant across time, all remaining variables exhibit time-series variation.18 Table A2 reports the country-by-country median values of the country-level explanatory variables. [Table 2 about here] To gain a basic understanding of how debt ratios and maturity structures are influenced by these variables, we compute the Pearson correlation coefficients for pairs of the dependent and independent variables. The results, reported in Table 3, suggest that

17

The set of country level independent variables are defined in Table A1, along with their data sources.

18

The corruption index prior to 1995 is taken as the 1988-1992 composite level, because compatible annual

data is not available prior to 1995.

18

the legal system, the tax system, and the suppliers of funds potentially influence the capital structure choice. In particular,

•

firms in more developed economies have lower debt ratios and more longterm debt;

•

common law is associated with lower leverage and more long-term debt;

•

low levels of corruption are associated with lower debt ratios and a greater use of long-term rather than short-term debt;

•

the existence of an explicit bankruptcy code is associated with higher debt ratios and a greater use of long-term debt;

•

firms in countries that have a higher tax preference for debt have higher debt ratios;

•

firms in countries with more bank deposits and larger domestic savings tend to have higher leverage and more short-term debt;

•

the existence of explicit deposit insurance is associated with more longterm debt;

•

the level of defined contribution pension fund assets is associated with lower leverage; and

•

the level of defined benefit pension fund assets is associated with the use of long-term debt.19

19

In unreported analysis we examined these correlations in a number of subsamples. Specifically, we

separately examine developed and developing economies, and we divide the sample between two time periods, 1991-1998 and 1999-2006. There are some differences between the subsamples. For example, we find that the correlation between leverage and common law is strong only in developed economies and

19

[Table 3 about here] To investigate whether these variables are likely to be subject to collinearity problems in our later regression analysis, we examine the correlations between the independent variables that are used in our analysis. From Table 3, we see that most variables are not highly correlated with each other, with some notable exceptions. In particular, the correlation between the economic development indicator variable and the corruption index is negative 76 percent.

V. Regression Analysis This section presents regressions that estimate the influence of country-level explanatory variables on capital structure choices controlling for firm- and industry-level characteristics. Our regressions are estimated with a General Methods of Moments (GMM) approach that accounts for the fact that the regression residuals are heteroskedastic and serially correlated across both firm and country level observations.20 taxes and deposit insurance are strongly correlated with leverage in only the initial sub-period. In addition, the size of the government bond market is negative and highly correlated with leverage in developing economies, but only in the initial sub-period. Deposit insurance is positively correlated with debt maturity in only the later sub-period while insurance penetration is positively correlated with debt maturity in both sub-periods, but only in developing economies. Finally, the size of the government bond market is negative and strongly correlated with debt maturity in the later sub-period but only in developed economies. 20

The regressions are performed on panel data where the residuals may be correlated across firms and/or

across country, and OLS standard errors can be biased. We use the ordinary least square (OLS) method with heteroscedastic / autocorrelation corrected (HAC) errors (Andrew, 1991) and clustered at the country level (Petersen (2008)). The HAC procedure accounts for the potential heteroscedasticity and autocorrelation at the firm level by deriving the t-statistics of estimated OLS coefficients from Generalized

20

A. The Determinants of Leverage Table 4 presents the results of the leverage regressions. 21 Column one reports the regression for the full sample, column two provides evidence for the sub-sample of developed economies only and column three the sub-sample of developing economies only. Columns four and five provide evidence for the sub-periods, 1991-1998 and 19992006, respectively, and Column six provides evidence for a sub-sample representing OECD countries for which pension fund bond/equity holding restriction information is available. Column seven provides evidence for a select sub-sample of OECD countries for which pension fund asset information is available. [Table 4 about here]

1. Firm Effects The top half of Table 4 reports the coefficient estimates of our firm-specific variables. These coefficient estimates indicate that leverage is positively related to asset tangibility and firm size and negatively related to profitability, and the market-to-book ratio. These results, which hold in the full sample as well as the sub-samples, are consistent with evidence on U.S. firms (Bradley, Jarrell, and Kim (1984), Titman and Wessels (1988)) and more recent international evidence (Rajan and Zingales (1995) and Booth, Aivazian, Demirguc-Kunt, and Maksimovic (2001), De Jong, Kabir, and Nguyen (2008)). These Methods of Moments (GMM) standard errors corrected for heteroscedasticity and auto-correlation. 21

The results are robust to the use of alternative proxies for the country’s legal system, corruption, taxation

and financial market development. Alternative proxies leave unaffected other estimated coefficients. In addition, results are substantially unchanged when we winsorize all variables at the 1 percent level.

21

results are also generally consistent with individual country leverage regressions that we report in Table A3. The coefficients for the market-to-book ratio have the same sign in all country regressions. Asset tangibility and size are positively related to leverage in 38 and 34 out of 39 countries, respectively. Finally, profitability is negatively related to leverage in 36 out of 39 countries.

2. Country Effects The lower half of Table 4 reports coefficient estimates for country variables.

The

regression for the full sample (Column 1) has an adjusted R-square of 0.1798 which is the same order of magnitude as the preliminary result reported previously, regressing leverage on firm-specific variables, industry and country fixed effects. These coefficient estimates indicate that leverage is positively related to economic development, but unrelated to both inflation and inflation volatility. Consistent with better investor protection leading to a greater use of equity financing, we find that corruption is associated with higher debt ratios, common law systems are associated with lower debt ratios and the existence of an explicit bankruptcy code is associated with higher debt ratios.22 In addition, we find that leverage is higher in countries where the tax

22

We also considered the possibility that in some countries regulatory barriers to entry might decrease the

risk of incumbent firms and thereby increase their debt capacities. To examine this in more detail, we considered regulatory variables introduced in Djankov, La Porta, Lopez de Silanes, and Shleifer (2002). However, because these variables are highly correlated with corruption and our common law indicator variable, we did not include them in the regression reported in Table 4. The high correlation between these entry barrier variables and common law and corruption, however, may partially explain why common law and corruption have such a strong effect on the capital structure choice.

22

gain from leverage is positive. This evidence contrasts with Booth, Aivazian, DemirgucKunt, and Maksimovic (2001) who do not find a significant relation between debt ratios and tax policy. This difference in results arises because of differences in both the sample countries and sample periods. The Booth, Aivazian, Demirguc-Kunt, and Maksimovic (2001) sample is mainly developing economies over the period 1980 to 1991. As we show, our evidence in favor of a tax effect comes from developed economies in a later time period. We find some support for the idea that suppliers of capital influence firm debt ratio choices. In particular, we find that leverage is higher in countries with deposit insurance, suggesting that the banking industry is important, but we do not find a significant relation between the size of the banking sector and debt ratios. In addition, we do not find a significant relation between leverage and the size of the insurance industry, the level of domestic savings or the size of the government bond market.23 However, in a select sample of OECD countries that report the level of pension fund assets we find that firms in countries with larger defined benefit pension funds have higher debt ratios and those with larger defined contribution pension funds have lower debt ratios. There are some significant differences between the subsamples. In particular, common law and the bankruptcy code are significant in the sample of developed 23

The weak result with respect to insurance penetration may be due to the lack of a clear prediction as to

the association between leverage and insurance penetration. Recognizing that life insurance incorporates both a savings (whole life and annuities) and an insurance (term insurance) product, we follow Beck and Webb (2003) and the suggestion of an anonymous referee, and proxy for the size of the insurance sector as a savings vehicle using insurance premiums/domestic savings. However, as with insurance penetration, we do not find a significant relation between leverage and insurance premiums/domestic savings.

23

economies, but not in the sample of developing economies; while deposit insurance and the size of the government bond market are important in developing economies, but not in developed economies. In addition, we find that the level of domestic savings and the size of the government bond market are significant in the 1991-1998 sub-period, but not in the 19992006 sub-period, while taxes and deposit insurance are important in the later time period, but not in the former period. The negative association between leverage and the size of the government bond market in the 1991-1998 sub-period suggests that there may be a fixed demand for fixed-income securities, so that government bonds crowd out corporate bond issues. The subsample analysis reveals that corruption is consistently associated with higher debt ratios in all subsamples. However, the common versus civil law distinction is less important in developing economies. On the other hand, deposit insurance is significant in the latter time period, reflecting, perhaps, an increase in the number of countries adopting explicit deposit insurance from 23 to 33. Taxes are significant in the sample of developed economies, but not in the sample of developing economies, and only in the later time period. This may be due to the observation that the influence of corporate taxes is likely to be weaker in countries where they are easier to evade. 24 In unreported regressions, we find that taxes are significant in a sample of below median tax evasion countries, but not in a sample of above median tax

24

The likelihood that the potential to avoid paying taxes influences the strength of the relationship between

taxes and leverage was pointed out to the authors by an anonymous referee

24

evasion countries. 25 With this index, our entire sample of developing economies is characterized as high tax evasion countries. In addition, we consider an alternative tax measure that considers only the tax treatment of dividends. As discussed in footnote 9, we estimate a dividend tax indicator variable that takes a value of one for countries with either a full dividend relief tax system or a full dividend imputation tax system and zero otherwise. With this measure we find that leverage is lower in countries that tax dividends less, and this result holds strongly in all subsamples and sub-periods. Finally, we find that the coefficients of inflation volatility, which are insignificant in about half the regressions, is significantly positive in the developed country subsample, the subsample of OECD countries for which pension fund asset information is available and in the total sample of countries in the 1999-2006 sub-period. We also find negative associations between leverage, the size of the banking sector and the existence of deposit insurance in the developed economy subsample.

These latter two findings are

inconsistent with our expectations, but appear to be driven by outliers.26

25

Based on the World Bank tax evasion index, World Economic Forum, Global Competitiveness Report

2001/2002. 26

In particular, the high leverage ratios of South Korean firms generate the relation between leverage and

inflation volatility, deposits and deposit insurances in both the developed economies subsample and the 1999-2006 sub-period. South Korea is characterised by high inflation volatility, a relatively small banking sector, the existence of deposit insurance and a relatively high level of domestic savings. After dropping South Korea, inflation volatility, the size of the banking sector and deposit insurance are all insignificantly related to leverage.

25

B. Determinants of Maturity Structure 1. Firm Effects Table 5 reports the results of the debt maturity structure regressions.27 These regressions are estimated on the full sample and the sub-samples as previously discussed. Column one reports the regression for the full sample, column two provides evidence for the subsample of developed economies only and column three the sub-sample of developing economies only. Columns four and five provide evidence for the sub-periods, 1991-1998 and 1999-2006, respectively. Column six provides evidence for sub-sample representing OECD countries for which pension fund asset information is available. [Table 5 about here] The coefficients of the firm-specific variables are largely consistent with prior research (Barclay and Smith (1995), Stohs and Mauer (1996), Guedes and Opler (1996), Demirguc-Kunt and Maksimovic (1999)) in the full sample and all sub-samples and subperiods. Long-term debt is used more by firms with greater asset tangibility, larger size and higher profits. However, in contrast to the findings in the U.S., we find that the market-to-book ratio is only weakly associated with debt maturity in the full sample and is unrelated to debt maturity in the developed economy subsample. Table A4 reports the results of the country-by-country debt maturity regressions. The most robust cross-sectional determinant of debt maturity is asset tangibility. With one exception, asset tangibility is significantly positively related to debt maturity

27

The results are robust to the use of alternative proxies for the country’s legal system, corruption, taxation

and financial market development. In addition, results are substantially unchanged when we winsorize all variables at the 1 percent level.

26

structure. On the other hand, we find cross-country variation in the signs of the estimated coefficients for profitability, firm size and the market-to-book ratio. Profitability is positively related to debt maturity structure in 25 (statistically significant in 15) out of 39 countries. Firm size is is positively related to debt maturity structure in 33 (statistically significant in 21) out of 39 countries and the market-to-book ratio is positively related to debt maturity structure in 28 out of 39 countries and is statistically insignificant in most countries. Indeed, this relation is significantly negative only in the U.S.28

2. Country Effects The estimates of the country level coefficients reveal that debt maturity is negatively related to the level of corruption, but positively related to the common law indicator variable, consistent with lower corruption and stronger investor protection encouraging the use of long-term debt financing. Also, the existence of an explicit bankruptcy code is associated with greater use of long-term debt. We find that debt maturity is positively related to the level of economic development. Overall our results with respect to the relation between maturity structure and country level characteristics are more robust than those reported for leverage. Consistent with the preferences of the suppliers of capital having an influence on the firms’ maturity structures, we find that debt maturity is strongly negatively related to 28

Prior literature also report mixed results. For example, Guedes and Opler (1996) report negative relations

for U.S. firms, while Stohs and Mauer (1996) find only mixed support for an inverse relationship for U.S. firms. Ozkan (2000) reports a positive relationship for U.K. firms. Outside the U.S. and the U.K., international evidence generally does not find significant relation between the two variables (Antoniou, Guney, and Paudyal (2006), Terra (2005)).

27

the amount of deposits in the country’s banking sector. This is in contrast to the negative, but insignificant banking sector result reported by Demirguc-Kunt and Maksimovic (1999). Further, we find that the level of domestic savings, measured as gross domestic saving over GDP, is negatively related to debt maturity. We also find that debt maturity is longer in countries with explicit deposit insurance reflecting the willingness of banks to lend longer-term debt when deposits are protected. In addition, debt maturity is shorter, the larger the government bond market. However, we find no reliable relation between maturity structure and the degree of insurance penetration. The weak result with respect to insurance penetration may be due to the lack of a clear prediction as to the association between maturity and insurance penetration.29 In general, although the results in all subsamples and sub-periods are similar, there are several exceptions. Inflation rate volatility is associated with shorter maturity in developed economies, but unrelated to maturity in developing economies. The size of the insurance industry is positively related to maturity in developing economies, but unrelated to maturity in developed economies. In addition, deposit insurance is associated with longer debt maturity only in the latter time period, possibly due to an increase in the number of countries adopting explicit deposit insurance from 23 to 33 over the full sample period. The size of the government bond market is negatively related to maturity only in developed economies and the relation is significant only in the latter time period. In addition, for a subsample of OECD countries we find that the level of defined benefit

29

As noted previously, we include a proxy for the size of the insurance sector as a savings vehicle using

insurance premiums/domestic savings. However, as with insurance penetration, we do not find a significant relation between maturity and insurance premiums/domestic savings.

28

pension fund assets is associated with greater use of long-term debt, while the level of defined contributions pension fund assets is unrelated to debt maturity. Finally, we find a significant positive relation between debt maturity and inflation in the developing economy subsample, in the earlier time period, and in the OECD subsample .30

C. Fixed-Effects and Cross-Sectional Estimates This section examines the extent to which the cross-sectional and time-series variation in our explanatory variables drive our results. Up to this point our emphasis has been on the cross-sectional variation in capital structures. However, the debt ratios in individual countries also vary from year to year, and some of that year to year variation may be explained by the year to year changes in our explanatory variables. To estimate the extent to which our results are generated from the cross section versus the time series we estimate both firm/country fixed-effects and Fama-MacBeth (1973) regressions. Specifically, we report fixed-effects leverage and maturity structure regressions in columns one and two of Table 6, respectively. The Fama-MacBeth (1973) leverage and maturity structure regressions with Newey-West corrected standard errors are reported in columns three and four, respectively. By sweeping out individual firm and country-effects, the fixed-effects regression estimates the extent to which the timeseries variation of our independent variables explains the time-series of capital structure choices. In contrast, the Fama-MacBeth (1973) regression estimate the regression model

30

The result that inflation is positively related to debt maturity in both developing economies and the

earlier time period is generated by the low inflation/short-term maturity characteristic of China. After dropping China, inflation is insignificantly related to maturity.

29

for each of the 16 years in sample period and then average the coefficients for all independent variables across the 16 years, isolating the cross-sectional determinants of capital structure. The coefficients and statistical significance of the independent variables are similar to those reported in the cross-sectional, time series regressions (Tables 4 and 5). [Table 6 about here] The regression estimates reported in Table 6 indicate that the relationships between financing choices and firm characteristics are significant in both the time-series and the cross-section, and are consistent with our earlier estimates. However, the results of the inflation and financial institution variables continue to be mixed. The results also show that several country variables, in particular corruption and deposit insurance, have significant effects on firm capital structure choices even though their time-serial variations are small.

D. Book Values and Financing Choices In this section we examine how the country variables affect the levels of short-term debt, long-term debt, and total debt relative to the asset values of the firms in our sample. We measure book leverage as the proportion of total debt to total assets of the firm, decomposing this measure into short-term debt to total assets and long-term debt to total assets. [Table 7 about here]

Table 7 presents the results of the book leverage regressions. Column one reports

30

the regression for total book leverage, column two provides evidence for the long-term debt ratio and column three provides evidence for the short-term debt ratio. As this table illustrates, the long-term debt ratio is higher in more developed economies, countries with common laws, lower corruption, explicit bankruptcy codes, relatively smaller banking sectors, deposit insurance, lower domestic savings, and smaller government bond issuances. By contrast, the short-term debt ratio is higher in less developed economies, countries not under common laws, countries with higher corruption, higher domestic savings, larger banking sectors, lack of deposit insurance, and higher domestic savings. Taken together, these results of the long- and short-term debt regressions are consistent and complementary with the results of the debt maturity ratio regressions in Table 5.31

VI. Summary and Conclusion At the outset, we described regression results that indicate that a firm’s capital structure is determined more by the country in which it is located than by its industry affiliation, suggesting that the institutional environment can have a profound effect on how firms are financed. Specifically, we find that a country’s legal and taxation system, level of 31

We also find that inflation rate volatility is associated with lower total debt ratio, which is in contrast to

the previous results in Table 4, where leverage is measured as total debt to market value. Note also that the size of the banking sector is associated with lower total debt ratio. However, this result is driven by the high leverage ratios of South Korean firms and relatively small banking sector of the South Korean economy. After dropping South Korea the size of the banking sector is insignificantly related to book leverage. Interestingly, the results presented in Table 7 indicate that the relationship between financing choices and tax is more important when leverage is defined relative to total assets rather than the market value of the firm.

31

corruption and the preferences of capital suppliers – banks and pension funds – explain a significant portion of the variation in leverage and debt maturity ratios. The effects of taxes on capital structure choices are consistent with theory. When the tax gain from leverage is positive, firms tilt their capital structures towards more debt. However, as we note below, the tax effect is not as strong and pervasive as other influences on capital structure. The legal environment also has an important influence on capital structure choices. Our strongest finding is that firms in countries that are viewed as more corrupt tend to be more levered and use more short-term debt. We also find that common law countries have lower leverage and use more long-term debt and that firms in countries with an explicit bankruptcy code have higher leverage and use relatively more long-term debt. We also provide evidence that suppliers of capital can influence how firms are financed. Most notably, the debt maturity structure of corporations in countries with larger banking sectors tend to be shorter, reflecting the preferences of banks to lend shortterm. However, controlling for the size of the banking sector, firms in countries with deposit insurance tend to have longer maturity debt, suggesting that deposit insurance in some way facilitates long term lending by banks. In contrast, our evidence of a relation between the size of the insurance sector and capital structure is very weak. However, we find that firms in countries with higher levels of defined contributions pension fund assets use relatively more equity, while firms in countries with higher levels of defined benefit pension fund assets use relatively more long-term debt. In addition, we find evidence that a larger government bond sector crowds out private debt capital in the developing countries, leading firms in these countries to borrow less. We do not, however, find an

32

effect of government borrowing on debt/value ratios of firms in developed countries, but we do find that firms in these countries tend to have shorter maturity debt when the government bond market is larger. While not all of our results hold across all subgroups and sub-periods, some of the results are quite strong and pervasive. This is particularly true in the debt maturity regressions where corruption, legal system and the size of the banking sector are very strong in all subsamples and sub-periods. Further, the bankruptcy code and domestic savings are also strongly related to debt maturity in all of the subsamples. Deposit insurance, while related to debt maturity in most subsamples, is insignificant in the 19911998 sub-period. In the leverage regressions the results depend more on subgroups and subperiods. For example, while we find that for the full sample, leverage is higher in countries where the tax gain from leverage is positive, we do not find a significant relation between the tax gain to leverage and debt ratios in the developing economies subsample and the tax effect is insignificant in the 1991-1998 sub-period. Likewise, the effect of both common law and bankruptcy code are insignificant if we restrict the sample to developing economies. However, the relationship between financial leverage and corruption is strong and significant in all subsamples. Although our emphasis has been on the effect of cross-country differences in institutions on corporate financial choices, our analysis may have implications for the

33

literature on how institutions can promote economic growth.32 Specifically, the fact that institutions influence how firms are financed may provide an indirect channel through which a country’s institutions affect economic growth. For example, there is reason to believe that if firms can raise more of their capital with equity and long-term debt, they will be better able to make longer-term investments, which may better promote economic growth. This suggests that an analysis of the relation between investment horizons and institutional structure offers an interesting avenue for future research.

32

Demirguc-Kunt and Maksimovic (1998), Levine and Zervos (1998), and Rajan and Zingales (1998) find

that, for a sample of developing and developed countries, the development of stock markets, bond markets and banks facilitate economic growth.

34

FIGURE 1 Median Leverage Ratio of Sample Firms (1991-2006) This figure plots the median leverage ratio across 39 different countries. The leverage ratio is measured as total debt over the market value of the firm. Total debt is defined to be the book value of current and long-term interest bearing debt. Market value of the firm is defined to be the market value of common equity plus book value of preferred stock plus total debt.

35

FIGURE 2 Median Long-term Debt Ratio of Sample Firms (1991-2006) This figure plots the median debt maturity ratio across 39 different countries. The debt maturity ratio is measured as long-term interest bearing debt over total debt. Total debt is defined to be the book value of current and long-term interest bearing debt.

36

TABLE 1 The Sample The table provides a description of the sample. The number of years that data is available for each country. The mean number of firms per year for each country. The median value of the proportion of firms represented in the sample for each country, by number of firms and market capitalization.

Country Australia Austria Belgium Brazil Canada Switzerland Chile China Germany Denmark Spain Finland France UK Greece Hong Kong Indonesia India Ireland Israel Italy Japan Korea Mexico Malaysia Netherlands Norway New Zealand Pakistan Peru Philippine Portugal Singapore Sweden Thailand Turkey Taiwan USA South Africa

Number of years of data used

Number of firms in the sample

Firm-years

16 16 16 16 16 16 16 13 16 16 16 16 16 16 16 16 16 15 16 13 16 16 16 16 16 16 16 16 16 14 16 16 16 16 16 15 16 16 16

1554 139 169 351 1865 274 158 1530 1011 208 223 175 1205 2861 321 939 295 637 109 181 343 4088 970 151 1011 280 266 134 114 74 188 110 628 447 481 201 1399 11119 558

8308 1144 1485 2591 10988 2656 1424 6827 9209 2123 2315 1684 9664 21785 2511 7108 2573 4388 880 949 2810 42611 6741 1230 7586 2612 1826 954 1061 491 1648 867 4111 3394 3457 1422 7051 77909 3699

37

Time series median value Number of firms in the Market capitalization of firms in sample/Total number of the sample/Stock market listed firms capitalization 0.50 0.75 0.59 0.46 0.34 0.72 0.41 0.57 0.90 0.64 0.17 0.95 0.87 0.67 0.64 0.62 0.62 0.07 0.89 0.15 0.73 0.97 0.40 0.51 0.71 0.73 0.77 0.49 0.11 0.20 0.53 0.67 0.76 0.86 0.60 0.40 0.68 0.81 0.53

0.79 0.61 0.49 0.87 0.79 0.77 0.58 0.44 0.69 0.88 0.48 0.83 0.73 0.62 0.53 0.85 0.70 0.39 0.49 0.36 0.55 0.86 0.72 0.92 0.86 0.88 0.92 0.94 0.42 0.55 0.81 0.61 0.82 0.91 0.73 0.74 0.74 0.81 0.78

TABLE 2 Summary Statistics The table provides the mean, standard deviation, median, minimum and maximum values of each variable. Leverage ratio is the ratio of total debt to market value of the firm. Total debt is defined to be the book value of short-term and long-term interest bearing debt. Market value of the firm is defined to be the market value of common equity plus book value of preferred stock plus total debt. Maturity structure ratio is the ratio of longterm debt to total debt. Tangible assets/total assets is the ratio of fixed assets to total assets, operating risk is measured as the absolute value of the annual change in ROA, ROA is the ratio of net income to total assets, firm size is measured as the natural logarithm of total assets and the market-to-book ratio is the ratio of market value of equity plus book value of total debt over total assets.

Country characteristic variables are

Development economy is a dummy variable equal to one when the country is classified as developed according to the World Bank classification based on countries’ gross national income levels. Inflation rate is the annual rate of change in a country’s CPI. Inflation rate volatility is the standard deviation of inflation rates from period t-4 to t. Corruption index is an index ranging from 0 to 10, with larger value indicating more severe corruption. Common law is a dummy variable equal to one when a country adopts the common law system. Bankruptcy code is a dummy variable equal to one if an insolvent firm is most likely to undergo a reorganization proceeding. Tax is an estimate of the miller tax ratio calculated using statutory tax rates. Deposits is the ratio of a country’s bank deposits to GDP. Deposit insurance is a dummy variable equal to one if bank deposits are insured by government. Domestic savings is the ratio of gross domestic saving to GDP. Insurance penetration is the value of a country’s total insurance premiums to GDP. Pension fund regulation index is an index of relative restrictions on debt and equity holdings of pension funds ranging from 1 to 4. Defined benefit pensions is the value of the country’s defined benefit pension fund assets to GDP. Defined contribution pensions is the value of the country’s defined contribution pension fund assets to GDP. Government bonds is the ratio of the value of domestically denominated government bonds to GDP.

38

Variables

N

Mean

Std Dev

Leverage ratio

264236

0.29

0.26

0.22

0.00

1.00

Maturity structure ratio

235874

0.53

0.34

0.57

0.00

1.00

Tangible assets/total assets

264236

0.33

0.24

0.29

0.00

0.97

ROA

264236

-0.13

0.98

0.02

-12.25

0.43

Log total assets

264236

19.76

4.21

19.82

-10.93

31.94

Market-to-book ratio

264236

2.50

6.87

1.51

-35.15

63.26

Developed economy

624

0.86

0.35

1.00

0.00

1.00

Inflation rate

624

0.03

0.05

0.02

-0.04

0.54

Inflation rate volatility

624

0.02

0.39

0.01

0.00

32.88

Corruption index

624

3.01

1.74

2.50

0.00

9.43

Common law

624

0.59

0.49

1.00

0.00

1.00

Bankruptcy code

624

0.68

0.47

1.00

0.00

1.00

Tax

624

0.23

0.15

0.28

-0.30

0.51

Deposits

624

0.93

0.57

0.67

0.13

2.46

Deposit insurance

624

0.87

0.34

1.00

0.00

1.00

Domestic savings

624

0.23

0.09

0.22

0.09

0.52

Insurance penetration

624

0.08

0.03

0.09

0.01

0.18

Pension fund regulation index

457

3.18

1.06

4.00

1.00

4.00

Defined benefit pensions

72

37.55

19.07

48.10

.08

71.33

Defined contribution pensions

72

24.72

18.41

29.86

.03

119.97

Government bonds

624

0.36

0.25

0.30

0.00

1.19

39

Median Minimum Maximum

TABLE 3 Correlation Matrix The table provides correlation matrix for our sample. Pearson correlation coefficients for all independent variables, leverage and debt maturity, together with each pairing of independent variables are presented. Variables are as defined in Table 2.

Leverage ratio Maturity structure ratio ROA Log total assets Market-to-book ratio Developed economy Inflation rate Inflation rate volatility Corruption index Common law Bankruptcy code Tax Deposits Deposit insurance Domestic savings Insurance penetration Pension fund regulation index Defined benefit pensions Defined contribution pensions Government bonds

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20]

[1] 1.000 0.062 0.033 0.182 -0.136 -0.075 0.021 0.014 0.165 -0.157 0.071 0.061 0.070 0.037 0.109 0.003 -0.034 0.148 -0.088 0.061

[2] 1.000 0.096 -0.030 0.012 0.139 0.001 -0.011 -0.210 0.173 0.095 0.036 -0.135 0.107 -0.229 -0.066 0.071 0.254 -0.089 -0.107

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

1.000 0.253 0.082 -0.058 0.013 0.005 0.066 -0.112 -0.060 0.014 0.045 -0.027 0.114 0.021 -0.068 -0.108 -0.082 0.072

1.000 -0.042 -0.122 -0.060 -0.028 0.313 -0.414 0.163 0.213 0.337 0.014 0.337 0.183 0.000 -0.463 -0.343 0.272

1.000 0.025 -0.001 -0.006 -0.041 0.035 -0.003 0.014 -0.027 0.009 -0.039 -0.010 0.029 0.065 0.046 -0.025

1.000 -0.374 -0.091 -0.757 0.054 0.283 -0.041 0.234 0.298 -0.401 0.317 0.431 0.158 0.075 0.158

1.000 0.340 0.328 -0.013 -0.045 -0.118 -0.325 -0.042 -0.117 -0.257 -0.160 0.385 0.272 -0.195

1.000 0.073 -0.041 -0.039 0.019 -0.044 -0.067 -0.003 -0.046 -0.080 -0.185 -0.108 -0.043

1.000 -0.293 0.089 0.080 -0.147 -0.130 0.323 -0.206 -0.192 -0.473 -0.295 0.015

1.000 -0.072 0.014 -0.315 -0.126 -0.349 -0.066 0.159 0.821 0.395 -0.467

40

TABLE 3 (continued) Correlation Matrix The table provides correlation matrix for our sample. Pearson correlation coefficients for all independent variables, leverage and debt maturity, together with each pairing of independent variables are presented. Variables are as defined in Table 2.

Bankruptcy code Tax Deposits Deposit insurance Domestic savings Insurance penetration Pension fund regulation index Defined benefit pensions Defined contribution pensions Government bonds

[11] [12] [13] [14] [15] [16] [17] [18] [19] [20]

[11] 1.000 0.344 0.088 0.521 -0.256 0.061 0.549 0.391 0.081 0.160

[12] 1.000 0.297 0.117 0.113 0.143 0.441 0.209 0.060 0.136

[13]

[14]

[15]

[16]

[17]

[18]

[19]

1.000 0.053 0.296 0.465 0.269 -0.622 -0.555 0.532

1.000 -0.406 0.102 0.337 0.323 0.141 0.234

1.000 0.013 -0.396 -0.686 -0.430 0.011

1.000 0.033 -0.176 0.128 0.281

1.000 0.225 -0.039 0.266

1.000 0.730 -0.591

1.000 -0.603

41

TABLE 4 Leverage, Firm and Country Level Determinants This table presents regressions of leverage on both firm and country level variables, as defined in Table 2. All regressions include dummy variables for industry (two digit SIC codes). The sample is divided between developed and developing economies as defined by the developed economy indicator variable, a sample of OECD member countries for which pension fund bond/equity holding restriction information is available and a select sample of OECD countries for which pension fund asset information is available. The sample is split into two sub-samples, 1991-1998 and 1999-2006. This table also reports the adjusted R-squared and number of firm-year observations. Standard errors are robust to clustering within country over time. T-statistics are given in parentheses.

42

Dependent variable: Total debt/Market value of the firm Full Developed Sample Sample Economies Independent variable (1) (2)

Developing Economy (3)

1991-1998 (4)

1999-2006 (5)

OECD Select OECD (6) (7)

Firm Factors: Tangible assets/total assets ROA Log total assets Market-to-book ratio

0.2178

0.2274

0.1171

0.1738

0.2359

0.2176

0.2221

(28.68)***

(25.93)***

(11.39)***

(19.32)***

(25.76)***

(25.65)***

(15.28)***

-0.0201

-0.0144

-0.2268

-0.1737

-0.0117

-0.0159

-0.0074

(-3.77)***

(-3.04)***

(-8.44)***

(-3.63)***

(-2.41)***

(-3.12)***

(-1.76)***

0.0065

0.0057

0.0109

0.0070

0.0060

0.0056

0.0029

(7.97)***

(7.09)***

(7.82)***

(4.51)***

(7.11)***

(7.05)***

(5.20)***

-0.0081

-0.0075

-0.0096

-0.0110

-0.0071

-0.0076

-0.0068

(-18.04)***

(-16.98)***

(-10.45)***

(-12.69)***

(-18.62)***

(-16.97)***

(-13.89)***

Country factors: Developed economy Inflation rate Inflation rate volatility Corruption index Common law Bankruptcy code Tax Deposits/GDP Deposit insurance

0.1006

0.1075

0.1071

0.1069

0.1361

(6.20)***

(4.42)***

(5.80)***

(4.53)***

(2.28)***

-0.0363

-0.2222

0.0917

-0.0674

0.0650

-0.0958

-0.6182

(-0.66)

(-0.92)

(1.75)*

(-0.58)

(1.17)

(-1.35)

(-1.47)

0.0043

1.1850

0.0033

0.0029

0.3570

0.0093

1.2617

(0.74)

(2.04)**

(0.57)

(0.41)

(2.29)***

(1.51)

(2.84)***

0.0222

0.0232

0.0261

0.0255

0.0183

0.0240

0.0092

(6.89)***

(6.28)***

(5.12)***

(5.69)***

(4.71)***

(6.60)***

(1.98)**

-0.0330

-0.0477

0.0430

-0.0537

-0.0266

-0.0410

-0.2615

(-3.91)***

(-5.43)***

(0.98)

(-4.46)***

(-2.61)***

(-4.52)***

(-7.23)***

0.0113

0.0097

-0.0136

0.0197

0.0068

0.0123

0.0378

(3.51)***

(2.87)***

(-1.51)

(4.16)***

(2.13)**

(2.66)***

(2.32)**

0.0654

0.1005

-0.0778

-0.0268

0.1790

0.1176

0.1295

(2.22)**

(3.05)***

(-1.01)

(-0.78)

(5.03)***

(3.55)***

(1.70)*

0.0004

-0.0198

-0.0057

-0.0178

0.0060

-0.0060

-0.0298

(0.05)

(-2.25)**

(-0.08)

(-1.13)

(0.72)

(-0.55)

(-1.48)

0.0069

-0.0066

0.0400

-0.0016

0.0116

0.0082

-0.0067

(2.36)***

(-2.15)**

(4.81)***

(-0.29)

(3.31)***

(1.85)***

(-0.59)

Domestic savings

0.0044

0.0008

-0.0102

0.0178

0.0007

0.0002

-0.0181

(1.25)

(0.19)

(-1.34)

(2.85)***

(0.22)

(0.04)

(-2.23)***

Insurance penetration

-0.0007

0.0930

0.1373

0.2201

-0.0645

0.0007

1.6296

(-0.01)

(0.71)

(0.42)

(1.14)

(-0.49)

(0.01)

(2.13)**

-0.0350

-0.0238

-0.1942

-0.1212

-0.0345

-0.0389

-0.2254

(-1.60)

(-1.07)

(-3.37)***

(-4.19)***

(-1.18)

(-1.72)***

(-5.37)***

Government bonds Pension fund regulation index

-0.0068 (-1.78)***

Defined benefit pensions

0.0032 (3.35)***

Defined contribution pensions

-0.0024 (-2.20)**

Number of observations

250668

218847

31821

87515

163153

232656

112722

Adjusted R-square

0.1798

0.1959

0.1689

0.1958

0.1932

0.1862

0.1891

*,**, and ***, significant at the 10, 5 and 1 percent level, respectively.

43

TABLE 5 Debt Maturity Structure, Firm and Country Level Determinants This table presents regressions of debt maturity on both firm and country level variables, as defined in Table 2. All regressions include dummy variables for industry (two digit SIC codes). The sample is dividend between developed and developing economies as defined by the developed economy indicator variable, a sample of OECD member countries for which pension fund bond/equity holding restriction information is available and a select sample of OECD countries for which pension fund asset information is available. This table also reports the adjusted R-squared and number of firm-year observations. Standard errors are robust to clustering within country over time. T-statistics are given in parentheses.

44

Dependent variable: Long-term debt/Total debt

Sample Independent variable

Full Sample

Developed Economy

Developing Economy

1991-1998

1999-2006

Select OECD Sample

(1)

(2)

(3)

(4)

(5)

(6)

Firm Factors: Tangible assets/total assets ROA Log total assets Market-to-book ratio

0.2648

0.2707

0.2990

0.2704

0.2659

0.2699

(46.94)***

(44.70)***

(23.77)***

(37.8)***

(37.34)***

(39.22)***

0.0765

0.0800

0.0650

0.0701

0.0733

0.0811

(24.22)***

(26.65)***

(4.30)***

(7.12)***

(22.36)***

(46.43)***

0.0139

0.0152

0.0157

0.0106

0.0149

0.0168

(23.32)***

(26.44)***

(12.12)***

(16.58)***

(20.74)***

(24.57)***

0.0005

0.0001

0.0034

0.0004

0.0003

-0.0038

(1.61)

(0.44)

(3.67)***

(0.83)

(1.23)

(-1.55)

0.1434

0.1296

0.0796 (2.09)**

Country factors: Developed economy

0.1422 (8.94)***

Inflation rate Inflation rate volatility Corruption index Common law Bankruptcy code Deposits/GDP Deposit insurance Domestic savings Insurance penetration Government bonds

(5.23)***

(7.12)***

0.0883

0.2335

0.1641

0.2414

0.0047

0.9699

(1.31)

(0.97)

(3.17)***

(2.09)**

(0.09)

(2.54)***

0.0046

-3.0938

0.0054

0.0002

-0.2641

-1.6375

(0.70)

(-5.00)***

(0.90)

(0.04)

(-1.73)*

(-4.65)***

-0.0352

-0.0335

-0.0145

-0.0321

-0.0386

-0.0043

(-11.71)***

(-9.57)***

(-4.03)***

(-8.45)***

(-9.81)***

(-1.81)*

0.0749

0.0742

0.1821

0.0977

0.0736

0.1424

(8.08)***

(7.54)***

(7.55)***

(6.85)***

(7.23)***

(5.12)***

0.0460

0.0489

0.0096

0.0599

0.0425

0.0258

(15.68)***

(14.50)***

(2.19)***

(9.90)***

(14.92)***

(2.84)***

-0.1245

-0.1176

-0.2805

-0.0968

-0.1320

-0.1414

(-16.33)***

(-13.43)***

(-5.73)***

(-7.81)***

(-15.32)***

(-8.35)***

0.0147

0.0147

0.0433

-0.0089

0.0283

0.0110

(4.33)***

(3.87)***

(8.98)***

(-1.61)

(8.73)***

(1.36)

-0.0335

-0.0419

-0.0201

-0.0332

-0.0353

-0.0116

(-11.76)***

(-12.75)***

(-4.25)***

(-6.02)***

(-11.75)***

(-1.69)*

0.1272

-0.1039

0.6509

0.1880

0.1398

0.6295

(1.06)

(-0.79)

(3.40)***

(0.88)

(0.91)

(1.09)

-0.0814

-0.1301

0.0205

0.0143

-0.0840

-0.0399

(-4.74)***

(-8.11)***

(0.64)

(0.53)

(-4.78)***

(-1.19)

Defined benefit pensions

0.0009 (1.73)*

Defined contribution pensions

-0.0006 (-0.91)

Number of observations

224527

194976

29551

81539

142988

97635

Adjusted R-square

0.2189

0.2041

0.2301

0.2068

0.2222

0.2405

*,**, and ***, significant at the 10, 5 and 1 percent level, respectively.

45

TABLE 6 Sources of Variation in Leverage and Debt Maturity This table presents regressions of both leverage and debt maturity. The results for leverage are reported in Panel A. Panel B reports the results for debt maturity. In both panels, Column (1) reports a fixed effects regression for leverage. Column (3) reports a Fama-MacBeth regression for leverage. Columns (2) and (4) reports the corresponding results for maturity structure. All variables are as defined in Table 2. Industry dummy variables (two digit SIC codes) are included in Columns (3) and (4). This table also reports the adjusted R-squared and number of firm-year observations. T-statistics are given in parentheses.

46

Dependent variable: Independent variable: Firm factors: Tangible assets/Total assets ROA Log total assets Market-to-book ratio

Fixed effects Total debt/Market Long-term value of the firm debt/Total debt

Fama-MacBeth Total debt/Market Long-term value of the firm debt/Total debt

(1)

(2)

(3)

(4)

0.1610 (7.23)*** -0.0260 (-8.29)***

0.0846 (5.03)*** 0.0136 (5.56)***

0.1964 (17.83)*** -0.1346 (-3.06)***

0.2692 (62.65)*** 0.0684

0.0110 (4.25)*** -0.0032 (-5.17 )***

0.0044 (3.74)*** 0.0006 (2.92)***

0.0075 (14.00)*** -0.0085 (-14.16)***

0.1952

0.0999

0.1140 (6.44)*** -0.0012

(2.15)** -0.0050 (-0.93) 0.0388 (2.59)***

(2.12)** -0.0046 (-2.83)*** -0.0206 (-1.97)**

(13.49)*** 0.0143 (21.85)*** 0.0000 (-0.04)

Country factors: Developed economy Inflation rate Inflation rate volatility Corruption index Common law Bankruptcy code Tax Deposits Deposit insurance Domestic savings Insurance penetration Government bonds Number of observations Adjusted R-squared

(-0.02) 0.1039 (0.62) 0.0206 (7.89)*** -0.0336 (-6.23)*** 0.0184

0.1885 (1.28) 0.0293 (0.61) 0.0310 (2.78)*** -2.7265

-0.0123 (-0.55) -0.0051 (-1.01) 0.9638

(4.89)*** -0.0036 (-0.15) -0.0243 (-2.15)** -0.0005 (-0.17) 1.1488

(-1.62) 0.0368 (0.39) -0.0240 (-1.09) 251780 0.0757

(1.33) 0.0081 (0.05) -0.0777 (-5.42)*** 225437 0.1089

(4.14)*** 0.5433 (3.36)*** -0.0403 (-1.70)* 251780 0.1102

*,**, and ***, significant at the 10, 5 and 1 percent level, respectively.

47

0.1536 (12.95)*** -0.1211 (-0.74) -0.1100 (-0.91) -0.0363 (-15.76)*** 0.0954 (20.04)*** 0.0526 (16.63)***

-0.1089 (-9.40)*** 0.0074 (1.37) -4.3573 (-13.62)*** -0.2434 (-2.65)*** 0.0002 (0.01) 225437 0.1923

TABLE 7 Book Debt Ratios This table presents regressions for book leverage, defined as is the ratio of total debt to total assets. We further decompose the book leverage ratio into its long-term and short-term components. Column (1) reports the regression for book leverage, defined as is the ratio of total debt to total assets. Column (2) reports the regression for long-term debt ratio, defined as is the ratio of long-term debt to total assets. Column (3) reports the regression for short-term debt ratio, defined as is the ratio of short-term to total assets. All variables are as defined in Table 2. All regressions include dummy variables for industry (two digit SIC codes). This table also reports the adjusted R-squared and number of firm-year observations. parentheses.

48

T-statistics are given in

Dependent variable Independent variable Firm Factors: Tangible assets/Total assets ROA Log total assets Market-to-book ratio

Country factors: Developed economy Inflation rate Inflation rate volatility Corruption index Common law Bankruptcy code Tax Deposits/GDP Deposit insurance Domestic savings Insurance penetration Government Bonds Number of observations Adjusted R-square

Total debt/Total Assets (1)

Long-term debt/Total Assets (2)

Short-term debt/Total Assets (3)

0.2016 (33.82)***

0.1887 (46.46)***

0.0110 (3.35)***

-0.1000 (-27.77)*** 0.0037 (6.60)*** -0.0024 (-5.23)***

-0.0120 (-6.74)*** 0.0063 (23.65)*** -0.0008 (-3.52)***

-0.0643 (-28.23)*** -0.0019 (-4.49)*** -0.0010 (-4.68)***

0.0667 (6.61)*** -0.0181 (-0.46) -0.0073 (-3.89)*** 0.0124 (5.34)***

0.0748 (9.28)*** -0.0141 (-0.66) -0.0003 (-0.15) -0.0032 (-1.90)***

-0.0098 (-1.78)* -0.0487 (-1.59) -0.0037 (-1.62) 0.0169 (15.17)***

-0.0200 (-5.51)*** 0.0124 (6.92)*** 0.1121 (8.70)*** -0.0163 (-4.04)***

0.0214 (5.34)*** 0.0194 (14.51)***

-0.0353 (-12.79)*** -0.0085 (-7.58)***

-0.0402 (-11.55)***

0.0281 (10.06)***

0.0004 (0.27) 0.0054 (2.65)*** -0.0545 (-0.84) -0.0337 (-3.56)***

0.0043 (3.31)*** -0.0069 (-4.74)*** 0.0393 (0.77) -0.0306 (-4.01)***

-0.0053 (-4.30)*** 0.0118 (10.96)*** 0.0233 (0.63) -0.0011 (-0.19)

224527 0.1474

224527 0.1802

224527 0.1676

*,**, and ***, significant at the 10, 5 and 1 percent level, respectively.

49

TABLE A1 Definitions and Data Sources of Country Level Variables Variable Developed economy

Inflation rate

Description A zero or one dummy variable indicating whether the country is classified as developed according to the World Bank classification based on countries’ gross national income levels Annual rate of change on Consumer Price Index

Source World Development Indicators, World Bank

World Development Indicators, World Bank Inflation rate volatility World Development Standard deviation of inflation rates from period t-4 to t Indicators, World Bank Corruption index An index ranges from 0 to 10, with larger value indicating Corruption Perception more severe corruption Index, Transparency International Common law Treisman [2000] An zero or one dummy variable indicating whether a country adopts the common law system Bankruptcy code A proxy for the existence of an explicit bankruptcy code, Djankov, Hart, McLiesh, measured as a dummy variable equal to 1 if an insolvent and Shleifer (2008) firm is most likely to undergo a reorganization proceeding. Tax Estimate of the miller tax ratio equal to (1 - [(after all tax Price Waterhouse Coopers, value of $dividends)/( after all tax value of $interest)]) Doing Business calculated using statutory tax rates Deposits A proxy for the degree of financial intermediation of a International Financial country, measure as the country’s deposits (liquid liability) Statistics, International over GDP Monetary Fund Demirguc-Kunt, Deposit insurance Dummy variable equal to 1 if bank deposits are insured by Karacaovali, and Laeven government (2005) Domestic savings A proxy for the level of domestic savings measure as the International Financial country’s gross domestic saving over GDP. Statistics, International Monetary Fund Insurance penetration Value of total insurance premiums/GDP. Total insurance Swiss Reinsurance Company premium are the sum of life and non-life insurance Beck and Demirgüç-Kunt premiums. (2009) Pension fund An index of relative restrictions on debt and equity OECD, Survey of regulation index holdings of pension funds measured as the ratio of the Investment Regulation of proportional limit on equity holdings over the proportional Pension Funds limit on debt holdings, with larger values indicating tighter restriction on bond holdings. The index ranges from 1 to 4. Government bonds Value of domestically denominated government International Financial bonds/GDP Statistics, International Monetary Fund OECD Defined benefit Value of defined benefit pension fund assets/GDP. pensions Defined contribution OECD pensions Value of defined contribution pension fund assets/GDP

50

TABLE A2 Median Values of Country Level Dependent Variables The table provides the median value of country level the dependent variables, classified by country. Variables are as defined in Table 2 and Table A1

51

Country Australia Austria Belgium Brazil Canada Switzerland Chile China Germany Denmark Spain Finland France United Kingdom Greece Hong Kong, Indonesia India Ireland Israel Italy Japan Korea, Rep. Mexico Malaysia Netherlands Norway New Zealand Pakistan Peru Philippines Portugal Singapore Sweden Thailand Turkey Taiwan United States South Africa

Developed economy

Inflation rate

Inflation rate volatility

Corruption index

Common law

Bankruptcy code

Tax

1.00 1.00 1.00 0.00 1.00 1.00 0.00 0.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.00 0.00 1.00 1.00 1.00 1.00 1.00 0.00 0.00 1.00 1.00 1.00 0.00 0.00 0.00 1.00 1.00 1.00 0.00 0.00 1.00 1.00 0.00

0.03 0.02 0.02 0.08 0.02 0.01 0.05 0.02 0.02 0.02 0.03 0.01 0.02 0.03 0.04 0.02 0.09 0.06 0.03 0.04 0.03 0.00 0.04 0.10 0.03 0.02 0.02 0.02 0.08 0.04 0.07 0.03 0.02 0.02 0.04 0.54 0.01 0.03 0.07

0.01 0.01 0.01 0.10 0.01 0.01 0.02 0.07 0.01 0.00 0.01 0.01 0.00 0.01 0.02 0.02 0.03 0.03 0.01 0.02 0.01 0.01 0.02 0.08 0.01 0.01 0.01 0.01 0.02 0.04 0.02 0.01 0.01 0.01 0.01 0.05 0.01 0.01 0.02

1.40 2.35 2.90 6.35 1.03 1.10 2.85 6.60 1.94 0.50 3.35 0.40 3.00 1.52 5.25 2.25 8.10 7.20 2.50 2.95 5.25 2.90 5.76 6.70 4.90 1.05 1.25 0.60 7.80 5.90 7.40 3.65 0.82 0.80 6.80 6.40 4.45 2.40 5.30

1.00 0.00 0.00 0.00 1.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 0.00 1.00 0.00 1.00 1.00 1.00 0.00 0.00 0.00 0.00 1.00 0.00 0.00 1.00 1.00 0.00 0.00 0.00 1.00 0.00 1.00 0.00 0.00 1.00 1.00

0.00 0.00 1.00 0.00 1.00 1.00 0.00 0.00 0.00 0.00 1.00 1.00 1.00 0.00 0.00 0.00 1.00 0.00 1.00 0.00 1.00 1.00 1.00 1.00 0.00 0.00 1.00 0.00 0.00 1.00 1.00 1.00 0.00 0.00 1.00 1.00 1.00 1.00 0.00

0.00 0.01 -0.02 0.15 0.19 0.30 0.17 0.30 0.00 0.05 0.18 0.00 0.17 0.13 -0.18 0.16 0.30 0.35 0.13 0.36 0.00 0.35 0.28 0.04 0.28 0.35 0.00 0.00 0.45 0.30 0.33 0.18 0.26 -0.03 0.00 -0.17 0.03 0.35 0.30

Deposits / GDP

Deposit insurance

Domestic savings

Insurance penetration

Pension fund regulation index

Government bonds

Defined benefit pension

Defined contribution pension

0.60 0.82 0.78 0.37 0.75 1.25 0.43 0.33 0.70 0.53 0.65 0.48 0.63 0.90 0.55 1.92 0.39 0.38 0.70 0.74 0.53 1.94 0.49 0.23 1.12 0.99 0.50 0.82 0.30 0.21 0.47 0.88 0.98 0.40 0.94 0.32 0.24 0.65 0.50

0.00 1.00 1.00 1.00 1.00 1.00 1.00 0.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.00 1.00 1.00 1.00 0.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.00 0.00 1.00 1.00 1.00 0.00 1.00 1.00 1.00 1.00 1.00 0.00

0.23 0.25 0.24 0.18 0.23 0.28 0.26 0.42 0.22 0.25 0.23 0.26 0.20 0.16 0.10 0.31 0.31 0.23 0.36 0.17 0.22 0.27 0.35 0.22 0.43 0.27 0.31 0.23 0.16 0.18 0.15 0.17 0.47 0.24 0.34 0.17 0.14 0.16 0.19

0.04 0.02 0.04 0.00 0.03 0.07 0.02 0.01 0.03 0.04 0.02 0.07 0.06 0.09 0.01 0.03 0.01 0.01 0.07 0.03 0.03 0.09 0.09 0.01 0.02 0.05 0.02 0.02 0.00 0.00 0.01 0.02 0.03 0.04 0.01 0.00 0.04 0.04 0.11

2.00 1.00 2.00 1.00 2.00 2.00 1.00 2.00 2.00 2.00 3.00 3.00 3.00 3.00 3.00 . . . 3.00 3.00 4.00 4.00 1.00 1.00 1.00 4.00 4.00 4.00 . . . 2.00 . 1.00 . 4.00 . 4.00 2.00

0.09 0.47 0.81 0.32 0.28 0.11 0.28 0.15 0.28 0.57 0.32 0.37 0.39 0.33 0.74 0.09 0.21 0.20 0.36 0.00 0.80 0.60 0.11 0.15 0.34 0.42 0.12 0.27 0.36 0.02 0.35 0.40 0.40 0.39 0.02 0.21 0.13 0.29 0.46

. . . 12.36 52.52 . . . . 2.71 0.10 57.85 1.16 . . 5.71 . . . 24.11 0.46 10.79 0.50 3.09 . . 6.51 3.28 . . . 11.07 . . . 0.38 . 48.11 .

. . . 5.88 1.49 102.87 59.08 . . 26.92 6.32 . . . . 15.07 . . . 4.13 2.04 0.56 . 6.04 . . . 8.53 . 10.45 . 0.48 . . 4.72 0.19 . 32.24 .

52

TABLE A3 Pooled Firm-level Regressions of Leverage by Country The table presents the regression of leverage on firm level variables as defined in Table 2. The regression equation is estimated for each country using the pooled time-series and cross-sectional sample. All regressions include dummy variables for industry (two digit SIC codes). This table also reports the adjusted R-squared and number of firm-year observations. Standard errors are robust to clustering within firm over time. T-statistics are given in parentheses. Country Code Australia Austria Belgium Brazil Canada Switzerland Chile China Germany Denmark Spain Finland France

ROA -0.0061

Log total assets 0.0264

Market-to-book ratio -0.0072

No of observations/ Adjusted R-squared 8221

(2.11)**

(-1.02)

(9.49)***

(-9.11)***

0.1224

0.2481

-0.0735

0.0133

-0.0128

1093

(2.86)***

(-1.74) *

(2.78) ***

(-3.59)***

0.1129

Tangible assets/ Total Assets 0.0343

0.1931

-0.5328

0.016

-0.0132

1406

(3.61) ***

(-4.34)***

(5.31)***

(-3.73)***

0.1824

0.1930

-0.1741

-0.0010

-0.0103

2579

(3.09)***

(-2.81)***

(-0.31)

(-3.33)***

0.0814

0.0924

0.0070

0.0165

-0.0102

10806

(5.13)***

(0.78)

(4.05)***

(-13.46)***

0.1003

0.3178

-0.1171

0.0036

-0.0186

2583

(7.63) ***

(-2.14) **

(0.83)

(-5.97) ***

0.1926

0.0817

-0.1870

0.0181

-0.0091

1413

(1.56)

(-1.01)

(1.87)*

(-2.91)***

0.0840

0.0490

-0.1378

0.0423

-0.0073

6815

(2.10)**

(-5.87)***

(5.26)***

(-6.85)***

0.1221

0.3018

-0.0520

0.0056

-0.0111

8497

(9.41) ***

(-3.02)

(3.73)

(-9.43)

0.1106

0.2503

-0.0762

0.0003

-0.0193

2006

(4.27)***

(-1.74)*

(0.08)

(-6.15)***

0.1368

0.0767

-0.8292

0.0019

-0.0127

2025

(1.76)*

(-6.83) ***

(0.85)

(-3.57) ***

0.1600

0.4595

-0.2516

0.0055

-0.0149

1591

(7.57)***

(-2.68)***

(1.78) *

(-4.19)***

0.2493

0.3408

-0.1791

0.0081

-0.0134

9063

(10.19)***

(-3.48)***

(4.35)***

(-13.25)***

0.1624

0.2412

-0.0126

-0.0035

-0.0070

20741

(18.10) ***

(-1.93) *

(-2.81) ***

(-16.15) ***

0.1289

Greece

0.0335

-1.2120

0.0239

-0.0076

2471

(0.72)

(-8.47)***

(3.30)***

(-7.50)***

0.2250

Hong Kong

0.2220

-0.0512

0.0035

-0.0130

6493

United Kingdom

53

Indonesia India Ireland Israel Italy Japan Korea Mexico Malaysia Netherlands Norway New Zealand Pakistan Peru Philippines Portugal Singapore Sweden Thailand Turkey Taiwan United States South Africa

(7.69)***

(-4.03)***

(1.84)*

(-10.85)***

0.2525

-0.5076

0.0034

-0.0145

0.0982 2452

(4.55)***

(-5.01)***

(0.70)

(-5.10)***

0.1586

0.4351

-1.3180

-0.0052

-0.0192

4381

(11.03)***

(-13.25)***

(-1.16)

(-8.95)***

0.4328

0.2047

0.0234

0.0065

-0.0091

854

(3.49) ***

(1.67) *

(1.09)

(-3.74)***

0.1095

0.5941

-0.0212

0.0221

-0.0081

934

(7.69)***

(-0.54)

(1.76)*

(-3.06)**

0.3667

0.1041

-0.5515

0.0117

-0.0213

2644

(1.80) *

(-2.89)***

(2.81)***

(-6.68)***

0.1169

0.3655

-0.6499

0.0096

-0.0055

41763

(18.55)***

(-6.60)***

(4.70)***

(-6.05)***

0.1116

0.1765

-0.1977

0.0595

-0.0201

6734

(4.46)***

(-4.92)***

(7.65)***

(-6.03)***

0.1764

0.0821

-0.8375

0.0143

-0.0392

1224

(1.03)

(-4.14)***

(1.31)

(-3.27)***

0.2132

0.1179

-0.3592

0.0079

-0.0123

7072

(3.77)***

(-6.27)***

(3.76)***

(-7.05)***

0.1093

0.2754

-0.0994

0.0049

-0.0108

2482

(6.05)***

(-4.19)***

(1.55)

(-6.52)***

0.1594

0.5780

-0.0231

0.0049

-0.0145

1757

(19.03)***

(-0.90)

(1.61)

(-5.47)***

0.4568

0.1006

0.0022

0.0033

-0.0150

917

(-2.29)**

(1.68) *

(0.11)

(0.38)

0.3415

-1.5769

-0.0064

-0.0197

1045

(5.13)***

(-9.76)***

(-0.67)

(-3.67)***

0.4428

0.0169

-1.3765

0.0157

-0.0167

491

(0.15)

(-8.23)***

(1.17)

(-1.50)

0.2523

0.1738

-0.1420

0.0217

-0.0201

1523

(2.46) **

(-2.92)***

(5.16)***

(-4.46)***

0.1484

-0.2483

-0.7286

0.0113

-0.0301

840

(-3.15)***

(-2.66) ***

(1.67) *

(-4.99)***

0.1775

0.2632

-0.0783

0.0080

-0.0186

3973

(8.63)***

(-2.32)**

(2.39)**

(-6.50)***

0.1290

0.4990

0.0115

-0.0016

-0.0161

3189

(13.16)***

(0.79)

(-0.43)

(-8.47)***

0.3777

0.0853

-0.5986

0.0290

-0.0128

3403

(2.09)**

(-3.77)***

(3.80)***

(-3.28)***

0.1505

-0.0963

-0.5521

0.0114

-0.0063

1422

(-1.54)

(-5.16)***

(1.98)**

(-3.02)**

0.1925

0.0505

-0.8067

0.0395

-0.0411

7051

(1.90)*

(-10.40)***

(10.21)***

(-9.31)***

0.3186

0.2774

0.0052

0.0020

-0.0072

77132

(33.04)***

(2.66) ***

(3.19)***

(-38.41)***

0.1076

0.0780

-0.0442

-0.0110

-0.0073

3483

(-4.18) ***

0.0622

(2.39)** (-1.65)* (-4.35) *** *,**, and ***, significant at the 10, 5 and 1 percent level, respectively.

54

TABLE A4 Pooled Firm-level Regressions of Debt Maturity Structure by Country The table presents the regression of debt maturity on firm level variables as defined in Table 2. The regression equation is estimated for each country using the pooled timeseries and cross-sectional sample. All regressions include dummy variables for industry (two digit SIC codes). This table also reports the adjusted R-squared and number of firmyear observations. Standard errors are robust to clustering within firm over time. Tstatistics are given in parentheses. Country Code Australia Austria Belgium Brazil Canada Switzerland Chile China Germany Denmark Spain Finland France United Kingdom Greece Hong Kong

Tangible assets/ Total Assets 0.1720

Market-to-book ratio -0.0011

No of observations/ Adjusted R-squared 6292

0.0487

0.0344

(6.15)***

(3.15)***

(6.54)***

0.3808

(-0.72)

0.0991

-0.1207

-0.0006

0.0005

1025

(3.72) ***

(-1.65)

(-0.09)

(0.10)

0.0674

0.4061

0.0641

0.0167

0.0028

1370

(5.63)***

(0.50)

(3.84)***

(1.14)

0.1333

ROA

Log total assets

0.3386

0.1300

0.0217

0.0045

2551

(6.98)***

(4.17)***

(5.58)***

(1.99) **

0.1304

0.2135

0.0469

0.0282

-0.0018

8672

(8.97)***

(3.05)***

(5.70)***

(-1.49)

0.1144

0.3620

0.0296

-0.0016

-0.0055

2500

(7.95) ***

(0.43)

(-0.33)

(-1.58)

0.1005

0.3038

-0.1973

0.0440

0.0086

1301

(3.32)***

(-1.23)

(1.87)*

(1.47)

0.1269

0.3953

0.0590

0.0422

0.0006

6499

(8.65)***

(2.43)**

(2.35)**

(0.37)

0.1570

0.4253

0.0480

-0.0019

0.0021

7523

(12.06) ***

(1.76)*

(-1.06)

(1.52)

0.0846

0.4435

-0.0095

-0.0046

0.0031

1931

(7.00)***

(-0.30)**

(-1.07)

(0.85)

0.1098

0.3516

-0.1312

-0.0036

0.0059

1967

(6.46)***

(-1..00)

(-1.25)

(1.36)

0.0726

0.2129

-0.1358

0.0024

-0.0040

1551

(4.04)

(-2.27)**

(0.61)

(-1.22)

0.0512

0.3438

-0.0037

0.0062

0.0008

8972

(11.21)***

(-0.11)

(2.93)***

(0.65)

0.0612

0.3790

0.0243

0.0077

0.0003

18605

(20.40) ***

(2.14) **

(4.55) ***

(0.41)

0.0897

0.4844

0.4544

0.0157

-0.0012

2262

(7.06)***

(3.14) ***

(1.20)

(-0.71)

0.1062

0.3965

0.0127

0.0012

0.0019

5862

55

Indonesia India Ireland Israel Italy Japan Korea Mexico Malaysia Netherlands Norway New Zealand Pakistan Peru Philippines Portugal Singapore Sweden Thailand Turkey Taiwan United States South Africa

(10.76)***

(0.75)

(0.45)

(1.22)

0.3888

0.2320

0.0088

0.0028

0.0786 2280

(6.15)***

(4.72)***

(1.51)

(1.00)

0.0841

0.4336

0.0711

0.0208

0.0012

4286

(10.01)***

(0.87)

(3.00)***

(0.57)

0.1092

0.3475

0.0875

0.0210

0.0033

766

(4.67)***

(2.26) **

(2.67) ***

(0.95)

0.1232

0.3784

-0.0453

0.0283

-0.0008

750

(5.21)**

(-0.65)

(2.42)**

(-0.21)

0.1572

0.2976

0.2107

0.0088

-0.0002

2589

(5.44) ***

(2.78)***

(1.87)*

(-0.08)

0.0650

0.4001

0.1380

0.0118

0.0038

38813

(22.97)***

(3.91)***

(6.72)***

(4.97)***

0.0754

0.1406

-0.0917

0.0223

0.0049

6303

(4.07)***

(-3.18)**

(3.62)***

(2.14)**

0.0353

0.3308

0.3566

0.0531

0.0033

1159

(3.98)***

(3.97)***

(3.09)**

(0.38)

0.1765

0.2271

0.1534

0.0017

0.0010

6471

(6.02)***

(6.13)***

(0.60)

(0.50)

0.0362

0.4106

-0.1163

0.0181

0.0058

2270

(7.21)***

(-3.13)

(3.99)***

(2.37) **

0.1063

0.3817

0.1571

0.0082

-0.0025

1626

(9.03)***

(4.89)***

(2.71) ***

(-0.79)

0.1974

0.3828

0.0444

0.0022

0.0021

829 0.1088

(5.11)***

(1.57)

(0.43)

(0.33)

0.6196

-0.2254

0.0107

0.0055

1007

(7.57)***

(-1.54)

(0.93)

(1.57)

0.2628

0.3047

-0.5953

0.0709

0.0170

467

(3.86)***

(-3.26)***

(4.71)***

(2.69)***

0.2179

0.1777

0.0992

0.0188

0.0048

1257

(1.96)*

(1.61)

(2.94)***

(0.89)

0.0577

0.2685

0.1624

0.0195

-0.0041

824

(2.66) ***

(1.10)

(2.03)**

(-0.54)

0.0658

0.4891

0.0633

0.0093

0.0003

3712

(12.33)***

(2.61)***

(1.58)

(0.11)

0.1357

0.1782

0.0159

-0.0071

-0.0018

2777

(3.92) ***

(0.64)

(-2.46)**

(-0.60)

0.0347

0.3126

-0.0663

0.0413

0.0119

3271

(6.73)***

(-1.49)

(3.61)***

(4.63)***

0.1136

0.2459

-0.0509

0.0206

0.0088

1286

(3.12)***

(-0.73)

(2.83)***

(2.79) ***

0.0505

0.3771

0.1934

0.0452

0.0191

6564

(11.08)***

(3.19)***

(9.26)***

(4.19)***

0.1224

0.3169

0.0878

0.0176

-0.0010

66323

(35.85)***

(32.97)***

(24.42)***

(-3.38)***

0.1801

0.3702

-0.0239

-0.0070

0.0063

3065

(9.45)***

(-0.75)

(-2.97) ***

(3.36) ***

0.1025

*,**, and ***, significant at the 10, 5 and 1 percent level, respectively.

56

References: Andrew, D. W. K. “Heteroskedasticity and Autocorrelation Consistent Covariance Matrix Estimation.” Econometrica, 59 (1991), 817-858. Antoniou, A.; Y. Guney; and K. Paudyal. “Determinants of Corporate Capital Structure: Evidence from European Countries.” Working paper, University of Durham (2002). Auerbach, A. J. “Share Valuation and Corporate Equity Policy.” Journal of Public

Economics, 11 (1979), 291-305. Barclay, M. J., and C. W. Smith. “The Maturity Structure of Corporate Debt.” Journal of

Finance, 50 (1995), 609-631. Beck, T., and A. Demirgüç-Kunt. “Financial Institutions and Markets Across Countries and over Time: Data and Analysis.” World Bank Policy Research Working Paper No. 4943 (2009). Beck, T., and I. Webb. “Economic, Demographic, and Institutional Determinants of Life Insurance Consumption across Countries.” The World Bank Economic Review, 17 (2003), 51-88. Booth, L.; V. Aivazian; A. Demirguc-Kunt; and V. Maksimovic. “Capital Structures in Developing Countries.” Journal of Finance, 56 (2001), 87-130. Botero, J.; S. Djankov; R. La Porta; and A. Shleifer. “The Regulation of Labor.”

Quarterly Journal of Economics, 119 (2004), 1339-1382.

57

Bradley, M.; G. A. Jarrell; and H. Kim. “On the Existence of an Optimal Capital Structure.” Journal of Finance, 39 (1984), 857-878. Campbell, J., and Y. Hamao. “Changing Corporate Financing Structure and the Main Bank System in Japan.” In Japanese Main Bank System, M. Aoki and H. Patrick, eds. Oxford University Press (1995). Caprio, G.; L. Laeven; and R. Levine. “Governance and Bank Valuation.” Working paper, World Bank and University of Minnesota (2005). Claessens, S.; S. Djankov; and L. Klapper. “Resolution of Corporate Distress in East Asia.” Journal of Empirical Finance, 10 (2003), 199–216.

Claessens, S.; S. Djankov; and A. Mody. “Resolution of Financial Distress: An International Perspective on the Design of Bankruptcy Laws.” World Bank Institute Development Studies, Washington, D.C. (2001).

Claessens, S.; S. Djankov; and T. Nenova. “Corporate Risk around the World.” Working paper, World Bank, CEPR, and Harvard University (2001). Claessens, S., and L. Klapper. “Bankruptcy around the World: Explanations of Its Relative Use.” American Law and Economics Review, 7 (2005), 253–283. Davydenko, S, and J. Franks. “Do Bankruptcy Codes Matter? A Study of Defaults in France, Germany and the U.K.” Journal of Finance, 63 (2008), 565–608.

58

Degryse, H., and S. Ongena. “Distance, Lending Relationship and Competition.” Journal

of Finance, 55 (2005), 231–266. De Jong, A.; R. Rezaul Kabir; and T. Nguyen. “Capital Structure Around the World: the Roles of Firm- and Country-Specific Determinants.” Journal of Banking and Finance, 32 (2008), 1954–1969. Demirguc-Kunt, A., and V. Maksimovic. “Stock Market Development and Firm Financing Choices.” Word Bank Economic Review, 10 (1996), 341-369. Demirguc-Kunt, A., and V. Maksimovic. “Law, Finance and Firm Growth.” Journal of

Finance, 53 (1998), 2107-2137. Demirguc-Kunt, A., and V. Maksimovic. “Institutions, Financial Markets, and Firm Debt Maturity.” Journal of Financial Economics, 54 (1999), 295-336. Demirguc-Kunt, A., and V. Maksimovic. “Firms as Financial Intermediaries: Evidence from Trade Credit Data.” Working paper, World Bank and the University of Maryland (2001). Demirguc-Kunt, A.; B. Karacaovali; and L. Laeven. “Deposit Insurance Around the World : a Comprehensive Database.” Working paper, World Bank (2005). Diamond, D. “Seniority and Maturity of Debt Contracts.” Journal of Financial

Economics, 33 (1993), 341-368. Djankov, S.; R. La Porta; F. Lopez-de-Silanes; and A. Shleifer. “The Regulation of Entry.” Quarterly Journal of Economics, 117 (2002), 1-37.

59

Djankov, S.; R. La Porta; F. Lopez-de-Silanes; and A. Shleifer. “Courts.” Quarterly

Journal of Economics, 118 (2003), 453-517. Djankov, S.; O. Hart; C. McLiesh; and A. Shleifer. “Debt Enforcement around the World.” Journal of Political Economy, 116 (2008), 1105-1149. Djankov, S.; C. McLiesh; and A. Shleifer. “Private Credit in 129 Countries.” Journal of

Financial Economics, 84 (2007), 299-329. Fama, E. F., and J. D. MacBeth. “Risk, Return, and Equilibrium: Empirical Tests.”

Journal of Political Economy, 81 (1973), 607–636. Fisman, R. “Estimating the Value of Political Connections.” American Economic Review, 91 (2001), 1095-1102. Fisman, R., and I. Love. “Trade Credit, Financial Intermediary Development and Industry Growth.” Journal of Finance, 58 (2003), 353-374. Flannery, M. “Asymmetric Information and Risky Debt Maturity Choice.” Journal of

Finance, 41 (1986), 19-37. Gatward, P., and I. Sharpe. “Capital Structure Dynamics with Interrelated Adjustment: Australian Evidence.” Australian Journal of Management, 21 (1996), 89-112. Giannetti, M. “Do Better Institutions Mitigate Agency Problems?

Evidence from

Corporate Finance Choices.” Journal of Financial and Quantitative Analysis, 38 (2003), 185-212.

60

Graham, J. R. “Debt and the Marginal Tax Rate.” Journal of Financial Economics, 41 (1995), 41-73. Graham, J. R. “Taxes and Corporate Finance: A Review.” Review of Financial Studies, 16 (2003), 1075-1129. Guedes, J., and T. Opler. “The Determinants of the Maturity of Corporate Debt Issues.”

Journal of Finance, 51 (1996), 1809-1833. Himmelberg, C.; R. G. Hubbard; and D. Palia. “Understanding the Determinants of Managerial Ownership and the Link Between Ownership and Performance.” Journal of

Financial Economics, 53 (1999), 353-384. La Porta, R.; F. Lopez-De-Silanes; and A. Shleifer. “Corporate Ownership around the World.” Journal of Finance, 54 (1999), 471-518. La Porta, R.; F. Lopez-de-Silanes; A. Shleifer; and R. Vishny. “Legal Determinants of External Finance.” Journal of Finance, 52 (1997), 1131-1150. La Porta, R.; F. Lopez-de-Silanes; A. Shleifer; and R. Vishny. “Law and Finance.”

Journal of Political Economy, 106 (1998), 1113-1155. La Porta, R.; F. Lopez-de-Silanes; A. Shleifer; and R. Vishny. “The Quality of Government.” Journal of Law, Economics and Organization, 15 (1999), 222-279. La Porta, R.; F. Lopez-de-Silanes; A. Shleifer; and R. Vishny. “Agency Problems and Dividend Policies around the World.” Journal of Finance, 55 (2000), 1-33.

61

La Porta, R.; F. Lopez-de-Silanes; A. Shleifer; and R. Vishny. “Government Ownership of Banks.” Journal of Finance, 57 (2002), 265-301. La Porta, R.; F. Lopez-de-Silanes; A. Shleifer; and R. Vishny. Investor Protection and Corporate Valuation.” Journal of Finance, 57 (2002), 1147-1170. Levine, R., and S. Zervos. “Stock Markets, Banks and Economic Growth.” American

Economic Review, 88 (1998), 537-558. MacKay, P., and G. M. Phillips. “How Does Industry Affect Firm Financial Structure?”

Review of Financial Studies, 18 (2005), 1433-1466. Miguel, A., and J. Pindado. “Determinants of Capital Structure: New Evidence from Spanish Panel Data.” Journal of Corporate Finance, 7 (2001), 77–99. Miller, M. H. “Debt and Taxes.” Journal of Finance, 32 (1977), 261-275. Modigliani, F., and M. H. Miller. “The Cost of Capital, Corporate Finance, and the Theory of Investment.” American Economic Review, 48 (1958), 261-297. Modigliani, F., and M. H. Miller. “Corporate Income Taxes and the Cost of Capital: A Correction.” American Economic Review, 53 (1963), 433-443. Myers, S. “Capital Structure.” Journal of Economic Perspective, 15 (2001), 81-102. Myers, S., and N. Majluf. “Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have.” Journal of Financial Economics, 13 (1984), 187-221.

62

Ozkan, A. “An Empirical Analysis of Corporate Debt Maturity Structure.” European

Financial Management, 6 (2000), 197-212. Ozkan, A. “Determinants of Capital Structure and Adjustment to Long Run Targets: Evidence from UK Company Panel Data.” Journal of Business Finance and Accounting, 28 (2001), 175-198. Perotti, E., and K. Spier. “Capital Structure as a Bargaining Tool: The Role of Leverage in Contract Renegotiation.” American Economic Review, 83 (1993), 1131-1141. Petersen, M. A. “Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches.” Review of Financial Studies, 22 (2008), 435-480. Petersen, M. A., and R. Rajan. “The Benefits of Lending Relationship, Evidence from Small Business Data.” Journal of Finance, 49 (1994), 3–37. Rajan, R., and L. Zingales. “What Do We Know about Capital Structure? Some Evidence from International Data.” Journal of Finance, 50 (1995), 1421-1460. Rajan R., and L. Zingales. “Financial Development and Growth.” American Economic

Review, 88 (1998), 559-586. Sapienza, P. “The Effect of Government Ownership on Bank Lending.” Journal of

Financial Economics, 72 (2004), 357-384. Schmukler, S., and E. Vesperoni. “Globalization and Firms’ Financing Choices: Evidence from Emerging Economies.” Working paper, World Bank and International Monetary Fund (2001).

63

Shleifer, A., and R. Vishny. “The Grabbing Hand: Government Pathologies and their Cures.” Cambridge, MA: Harvard University Press (1998). Smith, C. W., and J. B. Warner. “On Financial Contracting: An Analysis of Bond Covenants.” Journal of Financial Economics, 7 (1979), 117-161. Stohs, M., and D. Mauer. “The Determinants of Corporate Debt Maturity Structure.”

Journal of Business, 69 (1996), 279-312. Titman, S., and R. Wessels. “The Determinants of Capital Structure Choice.” Journal of

Finance, 43 (1988), 1-19. Titman, S. “The Modigliani and Miller Theorem and Market Efficiency.” Financial

Management, 31 (2002), 101-115. Treisman, D. “The Causes of Corruption: A Cross National Study.” Journal of Public

Economics, 76 (2000), 399-457.

64